Plant Class I formins represent a unique family of cytoskeletal organizers that are at the same time integral membrane proteins. We have previously reported that the GFP fusion of the Arabidopsis formin AtFH1, which is considered to be the main housekeeping Class I formin expressed in majority of tissues according to the available gene expression data, also transiently localised to the tonoplast in the root transition zone at or around the time of major membrane rearrangements that generate the large central vacuole typical for differentiated plant cells (Oulehlová et al., 2019 – https://doi.org/10.1093/pcp/pcz102). It is currently unknown whether AtFH1 has an active role in tonoplast rearrangements or whether it transiently passes through the tonoplast while being sorted for vacuolar degradation. A newly prepared fusion of AtFH1 with a different, pH-insensitive fluorescent protein, as well as pharmacological treatment with concanamycin A, showed, in addition to known AtFH1-GFP localisation, also presence of AtFH1 within vacuolar lumen in the root elongation zone and mature tissues, supporting the degradation hypothesis. However, we also found quantitative, sometimes significant changes in vacuolar organization in mutants lacking AtFH1, using both established vacuolar morphology metrics and a newly developed method for quantitative evaluation of tonoplast topology, supporting a tonoplast-associated function of AtFH1. Therefore, depending on the exact developmental stage and conditions, both of the hypotheses may be correct.

The role of Arabidopsis Class I formin AtFH1 during vacuolar development – to use or to degrade?

Transposable Elements (TEs) are the major force in genome plasticity and gene evolution, with LTR retrotransposons (LTR-TEs) constituting the most abundant class. LTR-TEs are considered autonomous when they contain the coding sequence for the translation of all factors required for their replication and transposition, and nonautonomous if they require that such factors are provided in trans. While several autonomous LTR-TEs has been found mobile in Arabidopsis thaliana epigenetic mutants or epigenetic recombinant inbreed lines (epiRILs), non-autonomous LTR-TEs ha never observed mobile in real time experiments. In this study, we observe that the LTR_gypsy transposon ATGP2N mobilises in epiRILs and transpose with high specificity into ribosomal repeats. Notably, an identical homologous region is present in the LTRs of ATGP2N and the autonomous LTR-TE, suggesting that ATGP2N hijacks ATGP2 factors for its transposition. Our studies shed light on the transpositional dynamics of nonautonomous LTR elements and contribute to expanding our knowledge of LTR transposition mechanisms.

Mobilisation of a nonautonomous LTR-retrotransposon in Arabidopsis thaliana.

Transcription factors (TF) play a pivotal role in gene regulation; however, their biophysical properties remain poorly understood. Conventional techniques have provided relatively static perspectives of their action, limiting our understanding of their rapid (millisecond scale) dynamical behavior in cellula. Fortunately, over the past decade, novel microscopy techniques have emerged, enabling us to transcend the limitations of resolution and gain a comprehensive, nanoscale view of their in vivo biodynamics directly within the nucleus. To unravel and enhance the intranuclear diffusion of transcription factors, we have integrated molecular and microscopic approaches, facilitating a dynamic representation of their activity.

Investigation of transcription factor dynamics by super-resolution microscopy

In most plants, the proportion of transcribed DNA is a small portion of the genome, and this is particularly evident for species with large genomes, e.g. hexaploid wheat (Triticum aestivum), where the 85% of the genome is composed of non-expressed repetitive DNA elements, and only about 60% of the annotated genes are expressed. Indeed, it is known that the expression of large portions of plant genomes are blocked by transcriptional gene silencing (TGS) mechanisms, controlled by epigenetic marks such as DNA methylation. In the model plant Arabidopsis thaliana, the genome wide suppression of TGS activates the expression of silenced genes and mobilises repetitive elements, generating epigenetic and genetic variation and producing new phenotypes. However, attempts to exploit this variation in crop species has been so far largely unsuccessful, mostly due to the different and more complex epigenetic regulation of larger, and often polyploid, crop genomes. Therefore, we are developing a tool to directly screen in planta for conditions able to increase epigenetic variability via the suppression of TGS. For this purpose, wheat cv. ‘Fielder’ was transformed with a reporter GUS cassette containing DNA repeats, to attract DNA methylation and establish TGS on the transgenic locus. To prove that TGS was effectively silencing the reporter gene, we exposed transgenic seedlings to Zebularine, a cytidine analog that inhibits DNA methyltransferases and suppresses TGS, resulting in an increase in GUS expression. Ongoing and future work includes the use of this reporter line to test conditions, or breeding combinations, able to affect TGS in wheat, which can generate epigenetic and genetic variation. The exploitation of such diversity in breeding programs will potentially contribute to the improvement of wheat productivity and food sustainability.

Exploitation of silenced genes for wheat breeding

Out of the tens of thousands of edible species, only 2500 species of plants underwent any human selection, and only 250 species are currently considered domesticated. Before agriculture began in the Neolithic, humans in the Fertile Crescent exploited a large diversity of plants, collecting and cultivating them, but many were abandoned. 

While the occurrence of domestication and the development of agriculture are considered a pivotal point in human history, one significant question remains. What made certain wild plants more likely to be successfully domesticated? A potential answer to this question lies in mutation rate. Domestication relies on novel phenotypes, therefore if novel phenotypes arise more quickly in some species than others, humans may favour these species, continuing to grow them, and abandoning others.

Using transcriptomic data, the rate of mutation has been estimated for crop progenitors (those that were domesticated) and never-domesticated close relatives, which could, in theory, have been domesticated. In addition, a dN/dS analysis has been carried out to investigate selection across these species. We focused on two main groups of edible plants – legumes and cereals from the Fertile Crescent – including some which were likely managed during the Neolithic and then abandoned as a crop. 

Taking this alongside other factors that could have influenced the ‘domesticability’ of some species over others, our work opens a discussion about whether certain species are predisposed to succeed under certain conditions. It therefore has implications for using neo-domestication to ensure food security and for biodiversity response in a changing climate. 


Domesticate this, not that: the implications of mutation rate and selection on present-day agriculture

P7.16 
PlantAct: plant-soil solutions for climate challenges

Catherine A. Walsh1 , Heribert Hirt2

1Lancaster University, UK, 2King Abdullah University of Sciences and Technology, Saudi Arabia

Anthropogenic greenhouse gas (GHG) emissions have meant devastating impacts for our planet with associated climate change effects reaching crisis point. The recent surpassing of a 1.5C rise in global average temperature above pre-industrial levels as set by the 2015 Paris agreement, means an urgent need for action to prevent further temperature rises causing irreversible outcomes. Currently, agriculture contributes 25% of total GHG emissions, which suggests a conversion to more sustainable practices would significantly reduce GHG outputs. Despite driving climate change, global agriculture also suffers, with increasing frequencies of drought and flooding alongside soil degradation challenging food security. PlantAct is an international, interdisciplinary think tank tackling the problem, through applying plant and soil science solutions to address the intense difficulties facing agricultural production. Furthermore, PlantAct aims to connect with a range of disciplines to deliver resilient solutions to address nitrogen use efficiency, carbon sequestration and promote plant health whilst maintaining food security. A novel, holistic approach to our food system will not only relay environmental benefits but also foster a change in attitude towards more plant based diets to improve health and well-being. In doing so, PlantAct seeks to engage students, young scientists, and the next generation to help bring around the urgent overhaul required to reduce GHG emissions within the food system while providing carbon storage solutions to support the health of our planet into the future. 


PlantACT: plant-soil solutions for climate challenges

Birds can adaptively regulate body mass in response to the cost-benefit trade-off between risks of starvation and predation. Despite this, little is known about how birds actually monitor, detect, and interpret their body mass and the mechanisms employed to regulate or adjust body mass to match their current ecological conditions. Using zebra finches, we experimentally increased perceived mass via attachment of weighted backpacks and provided the birds with either ad libitum standard mixed-seed diet or supplementary high-fat to investigate (a) how birds assess their own body mass and (b) the mechanisms birds employ to adjust mass. In both experiments, independent of diet treatment, birds with backpacks rapidly lost mass and reduced overall activity, time spent in active behaviors and food intake. We found no evidence that mass loss was due to stress associated with backpack attachment - there was no difference in plasma corticosterone concentrations directly after rapid mass loss. We found evidence that birds likely interpret body mass via a physical mechanosensory pathway rather than a physiological pathway - rapid loss of mass was not linked to changes in plasma metabolites (glycerol or triglyceride concentrations). Our results suggest that the process of energy balance and mass regulation can involve a greater array of mechanisms than simply matching energy in (food consumed) to energy out. Zebra finches are able to force a negative energy balance through other, unidentified, mechanisms, even while maintaining high-fat dietary intake and reducing overall activity.

The enviable ability of zebra finches to get slim, even when eating a tasty high-fat diet

Laminarin, one of the main storage polysaccharides of brown algae and diatoms, was recently found to be an important molecule in the world’s oceans by accounting for up to 26 ± 17 % of the particulate organic carbon. While digestive enzymes breaking down laminarin (laminarinases) have been intensely studied in a number of adult marine invertebrates, data on larval stages remain scarce. Abundantly inhabiting kelp forests as ecosystem engineers, especially sea urchins are found in laminarin-rich areas, suggesting that laminarin digestion plays a significant role throughout their life cycle. Considering the high laminarin abundance in microalgae, i.e. the food source of echinoderm larvae, digestive degradation might be beneficial, if not crucial, not only in adult sea urchins but also for larval performance within plankton. 
To disentangle the ecological and physiological role of laminarin for sea urchin larvae we conducted perturbation experiments with different laminarin concentrations and food sources (laminarin-rich and -poor microalgae) to evaluate the impact onto larval nutrition and performance. We analyzed (I) growth, survival and settlement success, as well as (II) laminarinase activity and expression patterns during larval development of the painted sea urchin Lytechinus pictus in response to the feeding regimes. 
The results contribute to a better understanding of laminarin dependent digestive dynamics in the early life stages of echinoderms, potentially affecting the recruitment of these important ecosystem engineers. 


Sea urchin larval digestion - the role of laminarin for larval fitness

Human activities represent one of the most dramatic forms of environmental change, affecting global biodiversity. Anthropogenic noise has increased in the last century with detrimental effects for wildlife on both individual and population levels. Sound pollution can compromise animal heath by imposing chronic stress and damaging the auditory system, affect intra- and inter specific relationships by disrupting communication and masking the presence of predators.
Shrews produce high-pitched laryngeal calls (“twittering calls”), with unclear function. It has been suggested as rudimentary echo-orientation mechanism, communication, or response to stressful stimuli. In this study, we analysed greater white-toothed shrews (Crocidura russula), trapped along a gradient of anthropogenic noise with the goal to link twittering calls to exposure to environmental stressors, through the analysis of faecal glucocorticoid levels. We hypothesize that animals inhabiting noisier locations would experience higher levels of stress, produce more twittering calls, and reduce exploratory behaviour to balance stress demands.
We live trapped 25 shrews in 6 locations with different sound patterns and maintained them in captivity for 5 consecutive days, for repeated measurement of twittering calls, exploratory behaviour, and stress levels.
Results show that variation in noise levels influences the shrews’ stress response, independently of the average noise levels. Neither average noise levels nor its variation induced a change of vocalisations. While time in captivity resulted in a reduction of exploratory behaviour and twittering calls, we did not find evidence supporting a link between stress levels and twittering calls.


Noisy places, noisy animals? - Shrew responses to sound pollution

Many fastidious (or unculturable) pathogens, such as the Candidatus Liberibacter spp. associated with citrus greening (HLB) disease pose an immense threat to crop production and cause billions of dollars in losses worldwide. Unlike studies of model plant-pathogen systems, studies of fastidious crop pathogens are greatly hindered due to the unculturable, obligate lifestyle of the pathogens and the relative recalcitrance of the host crops to conventional genetic and laboratory evaluation approaches. To overcome these bottlenecks, we developed an innovative plant hairy root-based efficacy testing system suitable for faster pre-screening of diverse resistance and susceptibility genes and small molecule-based antimicrobials. Using this approach, we identified multiple novel genes, antimicrobial peptides, and chemicals that showed efficacy against Candidatus Liberibacter spp. Ongoing field trials with selected candidates are promising, with improved crop yield and health. One or more of these solutions could be further deployed as products for crop disease management.

Novel strategies to screen and evaluate genetic- and chemical-based solutions for crop improvement

How steeply metabolic rate scales with body mass has been investigated and debated for a century; however, little is known about scaling within the same individuals as they grow, let alone how environmental variability affects such ontogenetic scaling relationships. We quantified the ontogenetic (within-individual) scaling of metabolic and growth rates of juvenile rainbow trout (Oncorhynchus mykiss) at 10°C after different early-life temperature treatments: (1) 10°C throughout the experiment; (2) 14°C as eggs, then 10°C from hatching onwards; and (3) 10°C as eggs, then 14°C from hatching to completion of the yolk-sac stage, then back to 10°C (10°C being the optimal temperature and 14°C simulating an early-life warm-spell). The steepness (scaling exponent) of both metabolic and growth rate scaling varied significantly but modestly among individuals. No difference was found in the scaling of standard metabolic rate (SMR) between treatments, but maximum metabolic rate (MMR) and aerobic scope (AS) scaled shallower in the 14°C-yolk group, whereas growth rate scaled steeper in the 14°C-egg group. This shows that a warm-spell can have lasting and negative carryover effects on juvenile aerobic performance (MMR and AS) if warming occurs in the yolk-sac stage, and induce compensatory growth if warming occurs in the egg stage. Correlations between metabolic and growth scaling exponents for individual fish further indicated that compensatory growth did not come at the expense of an elevated SMR but traded off with MMR and AS. The specific ontogenetic timing of temperature variability thus affects ontogenetic scaling of metabolic and growth rates differently and independently.

Early-life temperature variability has life-stage-specific effects on later-life ontogenetic scaling of fish metabolic and growth rates

As Journal of Experimental Botany (JXB) editors, we often receive manuscripts focusing on phenotyping drought tolerance and plant responses to water deficit. Despite extensive research identifying multiple drought tolerance genes in recent decades, practical field application remains limited. We have observed that the quality of research in this field frequently suffers from flawed experimental designs, inconsistent terminology, overinterpretation of data, or unrealistic lab-to-field extrapolations. To tackle these challenges, the JXB editorial board established a working group to guide better experiment design, data interpretation, and result reporting, focusing on experiments performed in supposedly ‘controlled conditions’. Our recommendations include:
• Establishing clear, testable hypotheses and objectives moving away from vague goals like "improving drought tolerance".
• Utilizing precise, consensus-driven terminology to clearly communicate objectives and hypotheses. 
• Designing experiments that adequately account for the complexities of genotype-environment (GxE) interactions, by including sufficient biological replicates and conducting multiple experiments as well as measuring soil, plant and environmental parameters in parallel to monitor complex cross-interactions. 
• Considering the impact of whole-plant transpiration on soil-moisture and stress levels, stemming from interactions with pot sizes and soil-substrate, and acknowledging that plant responses to drought depend on, and also affect, their developmental stage.
These guidelines aim to enhance research quality, contributing vital knowledge to combat the growing threat of drought to agriculture.


Guidelines for designing and interpreting drought experiments in controlled conditions 

The development of multicellular organisms requires the integration of signals controlling and coordinating cell fates. This process relies on numerous regulatory events that impact both chromatin structure and transcription. Chromatin undergoes global modifications at specific times and within specific cell subsets to dictate distinct cellular programs. However, the mechanisms by which chromatin is modulated during development remains highly elusive.
Plants are ideal models for investigating cell-specific chromatin remodelling. Plant cells do not move, and their plastic development is the result of an exquisite interaction between environmental and intrinsic signals. How does this signal crosstalk guide chromatin orchestration?
Our findings, derived from the study of flowering time in Arabidopsis may bring some answers. One key function of the plant hormone Brassinosteroids is to regulate the vegetative to reproductive fate transition of the cells in flower primordium. BR signalling acts through inactivation of GSK3 kinases responsible for the modulation of other protein’s activity. Interestingly, in contrast to other hormones, BR do not undergo long distant transport. We discovered that the BR dependent GSK3s bind and phosphorylate a central component of chromatin machinery involved in establishing and maintaining transcriptional programs. This finding may help explaining how chromatin is orchestrated in specific cells to guide cell fate transitions. 


Chromatin Orchestration in Plant Development: Insights from Brassinosteroid Signaling and Flowering Time Control

SETBP1 haploinsufficiency disorder (SETBP1-HD) is a rare human genetic disorder (currently only 131 known cases) that affects development —particularly neurodevelopment. Individuals carry a hemizygous deletion or LoF in the SETBP1 region, which results in insufficient production of SET binding protein 1. Typically, SET binding protein 1 alters histone methylation to regulate the expression of target genes. When SETBP1 expression is low, key developmental genes are underexpressed, leading to developmental delay phenotypes. Symptoms include childhood hypotonia, motor development delay, intellectual and learning disabilities, anxiety, autism spectrum disorder, and attention-deficit/hyperactivity disorder. The aim of this study is to establish whether zebrafish represent a useful biomedical model for SETBP1-HD. Specifically, we deleted a single copy of SETBP1 via CRISPR transgenesis to investigate whether zebrafish SETBP1 hemizygosity induces the behavioral syndromes characteristic of human patients. Social and cognitive behaviours were assessed through larval behavioural assays; including thigmotaxis, startle response, and light dark response. A zebrafish model of SETBP1-HD will increase pathophysiological understanding of the disease and help to determine the molecular underpinnings of disease-linked variants, including Schinzel-Giedion syndrome and leukemia. Zebrafish represent a great model for high-throughput drug screening, which means that our transgenic SETBP1-HD lines may aid in the screening and development of potential SETBP1-HD therapies.

Modelling SETBP1 Haploinsufficiency Disorder (SETBP1-HD) in Zebrafish (Danio rerio)

In recent years, the COVID-19 virus pandemic has threatened public health globally. In several nations, the pandemic's aftermath is still affecting the human population. Numerous mutant strains of SARS CoV-2 spreading caused several waves of infection. These variants were linked to high infectivity or high mortality. But in the face of such assaults, human immune system developed a defence mechanism of its own against the infection to deal with cytokine storm, antibody neutralization escape and multiple organ failure. In order to boost the defence mechanism and SARS CoV-2 broadly being a respiratory disorder was targeted with multiple remedial strategies procured from traditional knowledge of medicinal plants. A survey was conducted to surmise the usage of such medicinal plants with therapeutic properties during the pandemic in a random population in West Bengal, India. Our data demonstrates the most often utilized raw traditional medicines. To deal with clinical manifestations associated with SARS CoV-2 (such as cough and cold) and boosting immunity, Tulsi (Ocimum tenuiflorum) was the most widely used plant which is easy to manage in kitchen gardens and have a high growth rate on the Indian subcontinent. This survey-based study is done to correlate with the efficacy of Tulsi in combating infectious diseases from common cough and cold to COVID-19. Regulated long-term use of these medicinal plants will also help avoid a number of prevalent illnesses and be advantageous for household healthcare. 


Immune Boosting Against SARS CoV-2 Infection: An Approach to Traditional Medicine

Climate change is impacting temperatures globally, altering the frequency, intensity, and extent of extreme events such as heatwaves and droughts. This modification poses a challenge to biodiversity worldwide and presents a particular threat to tropical biota, which already reside close to their thermal maxima. Fish play a crucial role in the diet of the Amazon, particularly for traditional communities that rely on them as their primary protein source. The Amazon boasts the largest diversity of freshwater fish globally, each exhibiting numerous physiological and behavioural adaptations to thrive in the warm and often hypoxic waters of the Amazon. Comprehending how rising temperatures affect Amazonian fish species is essential for policymakers tasked with crafting strategies to safeguard the bio and sociodiversity of the Amazon. Faced with increasing temperatures, fish respond by adjusting their physiology and behaviour until they reach their thermal limit. Traditional thermal tolerance tests conventionally use the loss of equilibrium as the final endpoint for determining thermal limits. However, well before this loss of equilibrium, fish undergo physiological and behavioural adjustments to cope with the changing conditions. Monitoring behavioural changes during critical thermal tolerance tests provides insights into how fish utilise physiological and behavioural adaptations in this progressively warmer environment. We will elucidate behavioural changes observed during thermal tolerance tests of various Amazonian species and explore their connection to physiological adaptations to the Amazonian environment. Additionally, we will consider how these changes can serve as early warning signs of thermal vulnerability for fish species.

Dealing with changes in an extreme world: how is living in tropical waters?

The session will address recent developments in the study of the cytoskeleton, especially the evolutionarily conserved actin and microtubule networks. We shall focus on common organizational principles and processes shared by major eukaryotic lineages, on the interplay between cytoskeletal dynamics and (intra)cellular motility, as well as on the role of the cytoskeleton in shaping cells and tissues.

**Session Organisers:** Dr. Fatima Cvrckova and Prof. Patrick Hussey

**Invited Speakers:** Dr. Marcus Braun, Prof. Robert Insall, Dr. Teije Middelkoop, Prof. Sabine Müller, Dr. Kateřina Schwarzerová and Prof. Michael Sixt

C3 - The Cytoskeleton Across Kingdoms

poster

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

If you see this page it means you are not logged in or registered.

If you feel like you see this message by mistake please make sure you are logged in with the email you registered with.

If you are not registered yet, you can do so [**here**](https://www.sebiology.org/events/seb-conference-prague-2024/registration2024.html). 

Please login or register

SEB Conference Prague 2024

SEB’s Annual Conference is celebrated for creating the perfect composition of research, new discoveries and collaborative connections. Be creative and innovative by exchanging ideas and knowledge with masters of biology from all over the world.

The plant genome is particularly dynamic, with both genetic and epigenetic elements contributing to plasticity and adaptation to environmental changes. This session will bring together studies covering the fundamental mechanisms contributing to (epi)genome plasticity, and how this understanding can be applied to plant responses. The session aims to look for integration of genetics and epigenetics studies, including molecular biology and genomic approaches, and determine how plant (epi)genome plasticity can contribute to improve plant adaptation in a changing environment.

**Session Organisers:** Dr. Robyn Emmerson and Dr. Marco Catoni

**Invited Speakers:** Filipe Borges, Associate Prof. Frank Johannes and Dr. Sara Lopez-Gomollon

P3 - Genomic and Epigenetic Plasticity in Plants

Maintaining energy balance is critical for the survival of organisms and is becoming increasingly difficult under environmental change. Trade-offs, which result as a disruption in energy supply or demand, are fundamental to biology and are often imposed to explain variation in traits related to fitness. Since energy is a universal currency, it is often assumed that its acquisition (e.g., feeding) and expenditure (e.g., metabolic rate) set the limits for which resources can be allocated towards fitness enhancing processes such as maintenance, growth, and reproduction. Under climate change, warmer and more variable habitats are expected to increase metabolic rates (and thereby the cost of living) while reducing food availability in the environment of many organisms. Despite strong theoretical foundations formed over the past 40 years, it remains unclear how common these energy trade-offs are in nature, and how they might be mediated – either via greater energy reserves or energy saving mechanisms (such as mitochondrial efficiency). This session invites talks and posters on variation in energy acquisition (e.g., feeding, energy stores), expenditure (e.g., metabolism), and allocation (e.g., maintenance, life-history traits) from subcellular to whole animal and population-level processes, to improve our understanding of energy trade-offs.

**Session Organisers:** Dr. Neal Dawson, Dr. Edward Ivimey-Cook and Amanda Pettersen

**Invited Speakers:** Dr. Florencia Camus, Prof. Neil B. Metcalfe, Prof. David Reznick and Dr. Elisa Thoral

This session has kindly been sponsored by 

**Conservation Physiology**

![](https://assets.underline.io/markdown_image/1/image/db65740fed66a5103263956550e7e282.png)

A16 - Tipping the Scales: Balancing Energy Acquisition Expenditure and Allocation in an Ever-Changing World

technical paper

The Open Animal Biology session covers the whole breadth of animal biology, and therefore will contain those talks and poster presentations which don’t fit into the specific Animal Biology sessions. As much as possible presentations will be organised together with similar subject groups/themes, each set of talks acting like its own small session. This is therefore an interesting and important part of the scientific programme, with high quality presentations from across the animal biological spectrum.

 Session Aims & Objectives
The session is open to all animal biology research areas, and from speakers of all levels of experience. We particularly encourage presentations by post-grads, post-docs, and early career scientists.

**Session Organiser:** Dr. Jack Thomson

A22 - Open Animal

Animals have the ability to respond physiologically and behaviourally to environmental change with survival often predicated on the interaction between the two. Behaviour is key for understanding how anthropogenic environmental changes affect the resilience of animal populations. The goal of this session is to focus on the interplay between behaviour and physiology and how this allows ectothermic and endothermic animals to respond to environmental change (e.g., temperature, desiccation, hypoxia, pollutants). The session will showcase the importance of multi-disciplinary research and the importance of a whole organism approach when studying environmental change.
 
Find out more about The Company of Biologists http://www.biologists.com.

**Session Organisers:** Dr. Suzie Currie and Dr. Kath Sloman

**Invited Speakers:** Dr. Alex Gunderson, Dr. Daiani Kochhann, Dr. Heidi MacLean, Dr. Danielle McDonald, Dr. Rachael Morgan and Dr. Natalie Pilakouta

This session has kindly been sponsored by 

**The Company of Biologists** 

![](https://assets.underline.io/markdown_image/1/image/04a29ffb3c7675ec2e5c23d87add3b5e.png)

**Conservation Physiology **

![](https://assets.underline.io/markdown_image/1/image/e1125333787457260327a7a2ec5c3496.png)

A8 - Links Between Physiology and Behaviour in a Changing World

Epigenetics pervades every aspect of plant biology. Understanding plant epigenetics has the potential to revolutionise crop breeding by unlocking previously unattainable traits and providing new tools to interfere with gene expression.  
Join us in this session to discuss the latest advances in epigenetics in model species and crops. We will explore the impact of epigenetics on plant development, environmental interactions as well as crop productivity, and examine the latest insights into the mechanisms of epigenetics.   
Overall, this session aims to facilitate discussions on how to better transfer our knowledge from model species to crops in ways that will allow us to manipulate the epigenome for improved food security and climate resilience. 

**Session Organisers:** Dr. Tamara Lechon Gomez and Dr. Hans-Wilhelm Nützmann

**Invited Speakers:** Associate Prof. Miguel de Lucas, Dr. Zdravko Lorković, Claudia Martinho, Iva Mozgová, Hana Šimková and Prof. Jurriaan Ton

SAB2 - Plant Epigenetics: From Models to Crops

This session will showcase different approaches used in the field of translational plant biodiversity. Translational biodiversity aims at elucidation of mechanisms underpinning the evolution of plant diversity, and making the generated know-how accessible and applicable to plant breeding and agricultural production. The session will discuss how plant biodiversity can be most efficiently used to improve future crops. We will address key new findings as well as challenges in translation of fundamental results into crop breeding and identify and discuss knowledge gaps and research needs. With the session we aim to establish translational plant biodiversity as a key new approach for crop improvement.

**Session Organisers:** Prof. Stanislav Kopriva, Prof. Maria von Korff Schmising and Prof. Nicolaus von Wirén

**Invited Speakers:** Professor Katherine Denby, Dr. Franziska Fichtner, Johannes Kromdijk, Prof. Laura Rossini, Professor Nils Stein and Guido van den Ackerveken

This session has kindly been sponsored by IPK Gatersleben, CEPLAS and Journal of Experimental Botany

P7 - Translational Plant Biodiversity

A typical automotive or industrial end equipment system consists a multitude of analog, power and mixed signal ICs. These ICs are fundamentally driven by similar underlying vectors such as increasing performance and capability while reducing size and cost and maintaining or enhancing reliability.
The talk will focus on key analog and power component careabouts and reliability mechanisms that are critical for scaling these components and also discuss advances in technology and component design that enable scaling while achieving the required technology reliability.

Analog Technology Scaling and Reliability Implications

Quantum Computing has continued to gain traction in having the potential to shape the future of computing. In this talk, I will discuss some of the latest developments in the field - from a technical as well as business viewpoint. This presentation will also share engineering advancements and challenges, not only in the processors powering the quantum computer but the entire system. As we move on the path towards quantum advantage and broader commercialization of this technology, the focus on longer term reliability of the key components will become even more critical and aspects related to superconducting qubit-based quantum computers will be highlighted. Finally, a vision of where we see the inevitable convergence of high performance classical computers (HPC) with quantum computers, in what is termed as quantum-centric supercomputing will be presented.

Utility-scale Quantum Computing - Advances and Future Challenges

In the last decade, we have witnessed an unprecedented revolution of artificial intelligence (AI) as well as various smart devices such as smart phones, AV/VR devices, smart vehicles and robots. Semiconductor is a key player in this revolution as they are at the heart of many inventions related to the computing devices. These transformative trends of AI and smart devices will continue to open up explosive growth opportunities for the semiconductor community. 

Innovative Technologies for Sustainable Future of Semiconductor Industry

Si-friendly HfO2-based ferroelectric devices including FeRAMs and FeFETs are promising for future integrated memory and AI applications with high energy efficiency. On the other hand, a variety of reliability issues can pose serious limitations on the practical applications. This presentation will discuss the promise and reliability of HfO2-based FeRAMs and FeFETs with an emphasis on ferroelectric film thickness scaling, and will also touch on a new AI computation using these devices, reservoir computing.

Hafnium-Oxide-Based Ferroelectric Devices for Low-Power Memory and AI Applications: Promises and Reliability Challenges

**Abstract:** A typical automotive or industrial end equipment system consists a multitude of analog, power and mixed signal ICs. These ICs are fundamentally driven by similar underlying vectors such as increasing performance and capability while reducing size and cost and maintaining or enhancing reliability.
The talk will focus on key analog and power component careabouts and reliability mechanisms that are critical for scaling these components and also discuss advances in technology and component design that enable scaling while achieving the required technology reliability.

![](https://assets.underline.io/markdown_image/1/image/be0364236260c884a607e6d0d7d33b9d.jpg)

**[Sameer](https://underline.io/speakers/193291-sameer-pendharkar)**, *Texas Instruments, USA*

Analog Technology Scaling and Reliability Implications, Sameer Pendharkar

keynote

#### Welcome to the 62nd IEEE International Reliability Physics Symposium (IRPS)! 
**The 2024 IEEE International Reliability Physics Symposium (IRPS) was held in Dallas, Texas from April 14 to 18, 2024.** 

[![](https://assets.underline.io/uploads/markdown_image/1/image/9cc005f94b4ba4a81918eb49e0a51057.png)](https://underline.io/events/446/reception)
[![](https://assets.underline.io/uploads/markdown_image/1/image/acae953ba70ae2a57eed7b0250d6dd7d.png)](https://www.irps.org/)
[![](https://assets.underline.io/uploads/markdown_image/1/image/e1806748f2fc69f825e596872a7afdfd.png)](https://drive.google.com/drive/folders/1SmhpLZD9swYxrfP0TYF6327QTIciy3tV?usp=drive_link)
 

**[IRPS 2024 eGuide](https://drive.google.com/file/d/1DHxkWbafodK_D2HlDi7lwRK7KatgBgBD/view?usp=drive_link)** 
Please save the zip file locally and click the "start.pdf" to view the file. 
 
**IRPS 2024 General Chair Welcome Message**

On behalf of the Management Committee of the IEEE International Reliability Physics Symposium (IRPS) and the IRPS Board of Directors, it gives me great pleasure to extend a warm welcome to all who are planning to attend the IRPS 2024, scheduled to take place from April 14-18, 2024, in Dallas, TX. 
Over its 62-year history, IRPS has solidified its position as the premiere conference for engineers and scientists to present new and original work in the field of microelectronics reliability. Drawing participants from the United States, Europe, Asia, and other parts of the world, IRPS aims to deepen our understanding of reliability of semiconductor devices, integrated circuits, and microelectronic assemblies through an improved understanding of both the physics of failure as well as the application environment. IRPS offers diverse opportunities for attendees to enhance their knowledge and understanding of all aspects of microelectronics reliability. It is also an outstanding chance to meet and network with reliability colleagues from around the world.

The IRPS 2024 technical activities will begin with two days of a total of twenty tutorial sessions covering a broad spectrum of emerging reliability topics beyond CMOS device, circuit and system reliability. These sessions are devoted to both newcomers as well as experts. Three reliability Year in Review (YIR) talks focusing on neuromorphic computing, dielectric breakdown in leading edge devices, and GaN technology will be provided on Monday after the tutorials.

The IRPS 2024 technical program will feature four distinguished keynote talks in the plenary sessions delivered by Prof. Shinichi Takagi from The University of Tokyo and Dr. Su Jin Ahn from Samsung on Tuesday, Dr. Rajeev Malik from IBM on Wednesday, and Dr. Sameer Pendharkar from Texas Instruments on Thursday. The technical program, organized into three parallel tracks, consists of contributed and seventeen invited papers from leaders in industry and academia. Additionally, two invited talks by recipients of paper awards from our sister conferences, European Symposium on Reliability of Electron Device Failure Physics and Analysis (ESREF) 2023 and the International Integrated Reliability Workshop (IIRW) 2023, will be given in the technical program. 

Tuesday evening will host ten workshops, offering an opportunity for in-depth, face-to-face discussions on emerging reliability challenges, including the three IRPS 2024 focus topics. A poster session will be held in the Austin Ranch outside the symposium building on Wednesday evening. Both events are sure to be lively. The IRPS 2024 and the International ESD Workshop (IEW) will be co-hosted with IEW registrants participating in both IRPS and IEW programs.

The IRPS 2024 will adopt a hybrid conference style, allowing virtual access to live oral presentations hosted by the Underline. Real-time streaming of all technical events (excluding Workshops and Posters) will be available in the US Central Timezone, providing virtual attendees with Q&A opportunities. For those virtual attendees who cannot watch the live presentations, the material will be available on demand to all attendees after the symposium. All authors will be presenting in person.

Finally, I would like to extend my extreme gratitude to our patrons, exhibitors, and all participants for their invaluable support, which makes IRPS successful. 

Please enjoy the IRPS 2024!

Koji Eriguchi, General Chair 
of 2024 IRPS Management Committee
 
**About IRPS** 
The IEEE International Reliability Physics Symposium (IRPS) is the premiere conference for engineers and scientists to present new and original work in the area of microelectronics reliability. Drawing participants from the United States, Europe, Asia, and all parts of the world, IRPS seeks to understand the reliability of semiconductor devices, integrated circuits, and microelectronic assemblies through an improved understanding of both the physics of failure as well as the application environment. IRPS provides numerous opportunities for attendees to increase their knowledge and understanding of all aspects of microelectronics reliability, and to meet and network with reliability colleagues from around the world.

It looks like you are not registered for this event. 

To access the site please register [**here**](https://www.irps.org/registration-fees). 

Please register!

IRPS 2024 Main Conference

The IEEE International Reliability Physics Symposium (IRPS) is the premiere conference for engineers and scientists to present new and original work in the area of microelectronics reliability. IRPS seeks to understand the reliability of semiconductor devices, integrated circuits, and microelectronic assemblies through an improved understanding of both the physics of failure as well as the application environment. 

**Abstract:** Quantum Computing has continued to gain traction in having the potential to shape the future of computing. In this talk, I will discuss some of the latest developments in the field - from a technical as well as business viewpoint. This presentation will also share engineering advancements and challenges, not only in the processors powering the quantum computer but the entire system. As we move on the path towards quantum advantage and broader commercialization of this technology, the focus on longer term reliability of the key components will become even more critical and aspects related to superconducting qubit-based quantum computers will be highlighted. Finally, a vision of where we see the inevitable convergence of high performance classical computers (HPC) with quantum computers, in what is termed as quantum-centric supercomputing will be presented.

![](https://assets.underline.io/markdown_image/1/image/e9c56e7a0a0eb092e28a2cfe4f9588a4.jpeg)

**[Rajeev Malik](https://underline.io/speakers/258219-rajeev)**, *IBM Quantum, USA*

Utility-scale Quantum Computing - Advances and Future Challenges, Rajeev Malik

**Abstract:** In the last decade, we have witnessed an unprecedented revolution of artificial intelligence (AI) as well as various smart devices such as smart phones, AV/VR devices, smart vehicles and robots. Semiconductor is a key player in this revolution as they are at the heart of many inventions related to the computing devices. These transformative trends of AI and smart devices will continue to open up explosive growth opportunities for the semiconductor community. However, there are compelling technical challenges to fully grasp its growth potential. It has become even more challenging to meet the requirements such as Power, Performance and Area (PPA) together with reliability, while continuously scaling the physical dimensions of semiconductor devices below 10 nm. To address these challenges, technical innovations in terms of device structures, materials, processes and equipment are essential. After reviewing the evolution of the semiconductor technologies over the past 50 years, this paper will identify today’s technical challenges and discuss promising innovative technologies for sustainable future. These technologies encompass not only logic and memory technologies such as DRAM, Flash and emerging memory technologies but also advanced packaging technologies. 
**[Su Jin Ahn](https://underline.io/speakers/166445-su-jin-ahn)**, *Samsung Semiconductor R&D Center, Samsung*

![](https://assets.underline.io/markdown_image/1/image/42212bdbafaf3fd09b004d6552b7a8ea.jpg)

Innovative Technologies for Sustainable Future of Semiconductor Industry, Su Jin Ahn

**Abstract:** Si-friendly HfO2-based ferroelectric devices including FeRAMs and FeFETs are promising for future integrated memory and AI applications with high energy efficiency. On the other hand, a variety of reliability issues can pose serious limitations on the practical applications. This presentation will discuss the promise and reliability of HfO2-based FeRAMs and FeFETs with an emphasis on ferroelectric film thickness scaling, and will also touch on a new AI computation using these devices, reservoir computing.
 
![](https://assets.underline.io/markdown_image/1/image/ab29450bd41346d24022c4e499ed75cd.jpg)
 
[**Shinichi Takagi**](https://underline.io/speakers/76926-shinichi-takagi), *The University of Tokyo, Japan*

Hafnium-Oxide-Based Ferroelectric Devices for Low-Power Memory and AI Applications: Promises and Reliability Challenges, Shinichi Takagi

Recent rapid progress in the creation of artificial intelligence (AI) systems has been driven in large part by innovations in architectures and algorithms for developing large scale artificial neural networks. As a consequence, it’s natural to ask what role abstract principles of intelligence — such as Bayes’ rule — might play in developing intelligent machines. In this talk, I will argue that there is a new way in which Bayes can be used in the context of AI, more akin to how it is used in cognitive science: providing an abstract description of how agents should solve certain problems and hence a tool for understanding their behavior. This new role is motivated in large part by the fact that we have succeeded in creating intelligent systems that we do not fully understand, making the problem for the machine learning researcher more closely parallel that of the cognitive scientist. I will talk about how this perspective can help us think about making machines with better informed priors about the world and give us insight into their behavior by directly creating cognitive models of neural networks.

Bayes in the age of intelligent machines

For nearly a decade now, the problem that has been top of mind for me is how we might enable AI systems to robustly optimize for what people want, and to avoid causing harm – from robots and self-driving cars, to assistive devices and deep brain stimulation, to theory and toy models, to large language models and now Gemini. In this talk, I’ll take the opportunity to share a bit about my journey in this space, what lessons I’ve learned, and how we’re approaching the safety and alignment of frontier models at Google DeepMind.

My Journey in AI Safety and Alignment

As capabilities of foundation models skyrocket, openness plummets. In this talk, I argue that open-source models are essential for the long-term goal of building a rigorous foundation for AI. Greater access—from API to open-weight to open-source—enables deeper forms of research. API access allows us to push the frontier of agents, and I will present our recent work on simulation and problem-solving agents. Open weights enables reproducible research on safety, interpretability, and more generally, “model forensics”. Open-source unlocks fundamental innovations in architectures, training procedures, and data curation methods. Of course, the key obstacle for building open-source models is the resources required (data, compute, and research/engineering). I will conclude with some promising directions that leverage the community that bring us closer to the vision of open-source foundation models.

Open-Source and Science in the Era of Foundation Models

NLP research made significant progress, and our community’s achievements are becoming deeply integrated in society. The recent paradigm shift due to rapid advances in Large Language Models (LLMs) offers immense potential, but also led NLP to become more homogeneous. In this talk, I will argue for the importance of embracing variation in research, which will lead to more innovation, and in turn, trust. I will give an overview of current challenges and show how they led to the loss of trust in our models. To counter this, I propose to embrace variation in three key areas: inputs to models, outputs of models and research itself. Embracing variation holistically will be crucial to move our field towards more trustworthy human-facing NLP.

Are LLMs Narrowing Our Horizon? Let’s Embrace Variation in NLP!

Large Language Models (LLMs) are on track to reverse what seemed like an inexorable shift of AI from explicit to tacit knowledge tasks. Trained as they are on everything ever written on the web, LLMs exhibit “approximate omniscience”–they can provide answers to all sorts of queries, but with nary a guarantee. This could herald a new era for knowledge-based AI systems–with LLMs taking the role of (blowhard?) experts. But first, we have to stop confusing the impressive style/form of the generated knowledge for correct/factual content, and resist the temptation to ascribe reasoning, planning, self-critiquing etc. powers to approximate retrieval by these n-gram models on steroids. We have to focus instead on LLM-Modulo techniques that complement the unfettered idea generation of LLMs with careful vetting by model-based verifiers (the models underlying which themselves can be teased out from LLMs in semi-automated fashion). In this talk, I will reify this vision and attendant caveats in the context of our ongoing work on understanding the role of LLMs in planning tasks.

Can LLMs Reason and Plan?

 Adapting a model trained on vast amounts of data to new tasks with limited labeled data has long been a challenging problem, and over the years, a diverse range of techniques have been explored. Effective model adaptation requires achieving high accuracy through task-specific specialization without forgetting previous learnings, robustly handling the high variance from limited task-relevant supervision, and doing so efficiently with minimal compute and memory overheads. Recently, large language models (LLMs) have demonstrated remarkable ease of adaptation to new tasks with just a few examples provided in context, without any explicit training for such a capability. Puzzled by this apparent success, many researchers have sought to explain why in-context learning (ICL) works, but we still have only an incomplete understanding. In this talk, we examine this emerging phenomenon and assess its potential to meet our longstanding model adaptation goals in terms of accuracy, robustness, and efficiency.

Does In-Context-Learning Offer the Best Tradeoff in Accuracy, Robustness, and Efficiency for Model Adaptation?

Large language models motivate an approach to meaning known as distributional semantics, that words mean what they do because of how they’re distributed in language. In this talk I will describe some evidence from my lab that suggests metrics from large language models do a good job of predicting behavioral and neural responses to some aspects of human language. I go on to describe some research that highlights important differences in meaning processing in humans and the ‘understanding’ displayed by language models. Discrepancies are particularly noteworthy in studies of joke comprehension.

Distributional Semantics: What do large language models have to say?

How can Large Language Models (LLMs) and modern NLP be used to increase the use and the documentation of Indigenous languages which are in danger of disappearing? First, I report on the development of high-quality translators for Indigenous languages by fine-tuning SOTA machine translators with tiny amounts of data, and discuss how to avoid some common pitfalls. Next, I present prototypes built with Indigenous communities aiming to stimulate and facilitate writing, using LLM models to create spell-checkers, next-word predictors, and similar tools. Finally, I discuss a future for documentation where dying languages are preserved as interactive language models.

Harnessing the Power of LLMs to Vitalize Indigenous Languages

Recent years have witnessed the rise of increasingly larger and more sophisticated language models (LMs) capable of performing every task imaginable, sometimes at (super)human level. In this talk, I will argue that there is still space for specialist models in today’s NLP landscape. Such models can be dramatically more efficient, inclusive, and explainable. I will focus on two examples, opinion summarization and crosslingual semantic parsing and show how these two seemingly unrelated tasks can be addressed by explicitly learning task-specific representations. I will show how such representations can be further structured to allow search and retrieval, evidence-based generation, and cross-lingual alignment. Finally, I will discuss why we need to to use LLMs for what they are good at and remove the need for them to do things that can be done much better by smaller models.

Prompting is not all you need! Or why Structure and Representations still matter in NLP

That the recent LLM breakthroughs are solely due to scaling is a myth. Many difficult research problems had to be solved to make models like GPT4 and Mixtral possible. One of those difficult research problems is data quality. Data quality is a great challenge for NLP researchers with many opportunities for innovation and impact on current generative AI developments. I will focus on two examples in my talk: quality data for training a highly multilingual language model and quality data for instruction tuning.

Quality Data for LLMs: Challenges and Opportunities for NLP

**Karen Spärck Jones Award Lecture** 
The advent of generative AI technology produces transformative impact on the content creation landscape, offering alternative approaches to produce diverse, good-quality content across media, thereby reshaping the ecosystems of online content creation and publishing, but also raising concerns about market over-saturation and the potential marginalization of human creativity. Our recent work introduces a competition model generalized from the Tullock contest to analyze the tension between human creators and generative AI. Our theory and simulations suggest that despite challenges, a stable equilibrium between human and AI-generated content is possible. Our work contributes to understanding the competitive dynamics in the content creation industry, offering insights into the future interplay between human creativity and technological advancements in generative AI.

Human vs. Generative AI in Content Creation Competition: Symbiosis or Conflict?

**Title:** Bayes in the age of intelligent machines **Abstract:** Recent rapid progress in the creation of artificial intelligence (AI) systems has been driven in large part by innovations in architectures and algorithms for developing large scale artificial neural networks. As a consequence, it’s natural to ask what role abstract principles of intelligence — such as Bayes’ rule — might play in developing intelligent machines. In this talk, I will argue that there is a new way in which Bayes can be used in the context of AI, more akin to how it is used in cognitive science: providing an abstract description of how agents should solve certain problems and hence a tool for understanding their behavior. This new role is motivated in large part by the fact that we have succeeded in creating intelligent systems that we do not fully understand, making the problem for the machine learning researcher more closely parallel that of the cognitive scientist. I will talk about how this perspective can help us think about making machines with better informed priors about the world and give us insight into their behavior by directly creating cognitive models of neural networks.

Keynote: Tom Griffiths

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

EMNLP 2024

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

**Title:** My Journey in AI Safety and Alignment **Abstract:** For nearly a decade now, the problem that has been top of mind for me is how we might enable AI systems to robustly optimize for what people want, and to avoid causing harm – from robots and self-driving cars, to assistive devices and deep brain stimulation, to theory and toy models, to large language models and now Gemini. In this talk, I’ll take the opportunity to share a bit about my journey in this space, what lessons I’ve learned, and how we’re approaching the safety and alignment of frontier models at Google DeepMind.

Keynote: Anca Dragan

**Title:** Open-Source and Science in the Era of Foundation Models **Abstract:** As capabilities of foundation models skyrocket, openness plummets. In this talk, I argue that open-source models are essential for the long-term goal of building a rigorous foundation for AI. Greater access—from API to open-weight to open-source—enables deeper forms of research. API access allows us to push the frontier of agents, and I will present our recent work on simulation and problem-solving agents. Open weights enables reproducible research on safety, interpretability, and more generally, “model forensics”. Open-source unlocks fundamental innovations in architectures, training procedures, and data curation methods. Of course, the key obstacle for building open-source models is the resources required (data, compute, and research/engineering). I will conclude with some promising directions that leverage the community that bring us closer to the vision of open-source foundation models.

Keynote: Percy Liang

**Title:** Are LLMs Narrowing Our Horizon? Let’s Embrace Variation in NLP! **Abstract:** NLP research made significant progress, and our community’s achievements are becoming deeply integrated in society. The recent paradigm shift due to rapid advances in Large Language Models (LLMs) offers immense potential, but also led NLP to become more homogeneous. In this talk, I will argue for the importance of embracing variation in research, which will lead to more innovation, and in turn, trust. I will give an overview of current challenges and show how they led to the loss of trust in our models. To counter this, I propose to embrace variation in three key areas: inputs to models, outputs of models and research itself. Embracing variation holistically will be crucial to move our field towards more trustworthy human-facing NLP.

Keynote 3: Barbara Plank

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

**Title:** Can LLMs Reason and Plan? **Abstract:** Large Language Models (LLMs) are on track to reverse what seemed like an inexorable shift of AI from explicit to tacit knowledge tasks. Trained as they are on everything ever written on the web, LLMs exhibit “approximate omniscience”–they can provide answers to all sorts of queries, but with nary a guarantee. This could herald a new era for knowledge-based AI systems–with LLMs taking the role of (blowhard?) experts. But first, we have to stop confusing the impressive style/form of the generated knowledge for correct/factual content, and resist the temptation to ascribe reasoning, planning, self-critiquing etc. powers to approximate retrieval by these n-gram models on steroids. We have to focus instead on LLM-Modulo techniques that complement the unfettered idea generation of LLMs with careful vetting by model-based verifiers (the models underlying which themselves can be teased out from LLMs in semi-automated fashion). In this talk, I will reify this vision and attendant caveats in the context of our ongoing work on understanding the role of LLMs in planning tasks.

Keynote 2: Subbarao Kambhampati

**Title:** Does In-Context-Learning Offer the Best Tradeoff in Accuracy, Robustness, and Efficiency for Model Adaptation? **Abstract:** Adapting a model trained on vast amounts of data to new tasks with limited labeled data has long been a challenging problem, and over the years, a diverse range of techniques have been explored. Effective model adaptation requires achieving high accuracy through task-specific specialization without forgetting previous learnings, robustly handling the high variance from limited task-relevant supervision, and doing so efficiently with minimal compute and memory overheads. Recently, large language models (LLMs) have demonstrated remarkable ease of adaptation to new tasks with just a few examples provided in context, without any explicit training for such a capability. Puzzled by this apparent success, many researchers have sought to explain why in-context learning (ICL) works, but we still have only an incomplete understanding. In this talk, we examine this emerging phenomenon and assess its potential to meet our longstanding model adaptation goals in terms of accuracy, robustness, and efficiency.

Keynote 1: Sunita Sarawagi

**Title:** Distributional Semantics: What do large language models have to say? **Abstract:** Large language models motivate an approach to meaning known as distributional semantics, that words mean what they do because of how they’re distributed in language. In this talk I will describe some evidence from my lab that suggests metrics from large language models do a good job of predicting behavioral and neural responses to some aspects of human language. I go on to describe some research that highlights important differences in meaning processing in humans and the ‘understanding’ displayed by language models. Discrepancies are particularly noteworthy in studies of joke comprehension.

Keynote Speaker: Seana Coulson

## Welcome to 2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics!
This year, the conference is in Mexico City. NAACL was actually already planned for Mexico City in 2021, but due to the pandemic the entire conference was moved online. This year, finally, we get to go! So it is my sincere pleasure to welcome you to Mexico City, whether in person or virtually. Having the conference in Mexico City is a good opportunity to emphasize that NAACL is our flagship conference for ACL members not only in North America but also in Central and South America, even though NAACL has been bearing “North” in its name. At this year’s conference, we have a theme to match, with a theme track on the Languages of Latin America to showcase the linguistic diversity of the region.
 
The opportunity to present at NAACL should not depend on a researcher’s travel budget, or their family status. This is why it is so important to make virtual participation at NAACL as good an experience as possible – but we want to also provide a good experience for in-person participants. As a community, we are still working out the best way to do that. This year at NAACL, we are trying out a big virtual poster session ahead of the conference, with the hope that this will make make for a lively and interactive experience. At the same time, we are reducing virtual oral presentations, which seem to be particularly tricky to make to work well. A big thanks to the NAACL program chairs and to Luciana Benotti for all their ideas and work to improve the virtual experience. And participants, virtual as well as in-person: Please let us know what worked for you and what didn’t, so we can continue to improve hybrid conferences. 
 
I have been lucky to work with many amazing people. Without their insight, dedication and patience, and without the many hours of work they put in, NAACL would not have been possible. A huge thank you to the program chairs Helena Gomez, Kevin Duh, and Steve Bethard – you are the best! 
 
Finally, I would like to thank all authors, invited speakers and panelists, area chairs and reviewers, the volunteers organizing and chairing sessions, and all attendees, in-person and virtual. Thank you for helping us make NAACL 2024 come to life. 
 
Welcome and hope you all enjoy the conference! 
Katrin Erk 
The University of Texas at Austin 
NAACL-2024 General Chair 
*You can read the full Welcome message in the [Conference Handbook (downloadable)](https://drive.google.com/file/d/1H1NvW0VASQjkSCYgw3mr-3l4yYDSxz6B/view?usp=sharing)*

To access the event page you need to register [**here.**](http://acl.swoogo.com/naacl2024) 
Your access to the event page is limited based on your registration type. If you registered for workshops only, you will gain access to full workshops content on the day of the workshops program.

NAACL 2024

2024 Annual Conference of the North American Chapter of the Association for Computational Linguistics

**Title:** Harnessing the Power of LLMs to Vitalize Indigenous Languages **Abstract:** How can Large Language Models (LLMs) and modern NLP be used to increase the use and the documentation of Indigenous languages which are in danger of disappearing? First, I report on the development of high-quality translators for Indigenous languages by fine-tuning SOTA machine translators with tiny amounts of data, and discuss how to avoid some common pitfalls. Next, I present prototypes built with Indigenous communities aiming to stimulate and facilitate writing, using LLM models to create spell-checkers, next-word predictors, and similar tools. Finally, I discuss a future for documentation where dying languages are preserved as interactive language models.

Keynote Speaker: Claudio Santos Pinhanez

**Prompting is *not* all you need! Or why Structure and Representations still matter in NLP** **Abstract:** Recent years have witnessed the rise of increasingly larger and more sophisticated language
models (LMs) capable of performing every task imaginable, sometimes at (super)human level. In this talk, I will argue that there is still space for specialist models in today’s NLP landscape. Such models can
be dramatically more efficient, inclusive, and explainable. I will focus on two examples, opinion summarization and crosslingual semantic parsing and show how these two seemingly unrelated tasks can be
addressed by explicitly learning task-specific representations. I will show how such representations can be further structured to allow search and retrieval, evidence-based generation, and cross-lingual alignment.
Finally, I will discuss why we need to to use LLMs for what they are good at and remove the need for them to do things that can be done much better by smaller models.

Keynote: Mirella Lapata

Welcome to the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Continuing its mission of expanding and involving the science community of all European countries, EACL has selected the Malta community for the 18th EACL. Considering the importance of physical interaction among researchers, the conference will be held at the Hotel Radisson Blu, St. Julians, in Malta, from 17 to 22 of March, 2024. The conference will also feature the possibility to participate virtually. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language. 
#### [Conference Handbook](https://drive.google.com/file/d/13Wn38q0yev6U3RTgnf4Tn5T9GuAuZ-xW/view?usp=sharing) *downloadable*

To access the EACL 2024 event page you need to register [**here**](https://acl.swoogo.com/EACL2024). 

EACL 2024

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

**Quality Data for LLMs: Challenges and Opportunities for NLP** **Abstract:** That the recent LLM breakthroughs are solely due to scaling is a myth. Many difficult research problems had to be solved to make models like GPT4 and Mixtral possible. One of those difficult research problems is data quality. Data quality is a great challenge for NLP researchers with many opportunities for innovation and impact on current generative AI developments. I will focus on two examples in my talk: quality data for training a highly multilingual language model and quality data for instruction tuning.

Keynote: Hinrich Schütze

**Human vs. Generative AI in Content Creation Competition: Symbiosis or Conflict?** (*Karen Spärck Jones Award Lecture*) **Abstract:** The advent of generative AI technology produces transformative impact on the content creation landscape, offering alternative approaches to produce diverse, good-quality content across media, thereby reshaping the ecosystems of online content creation and publishing, but also raising concerns about market over-saturation and the potential marginalization of human creativity. Our recent work introduces a competition model generalized from the Tullock contest to analyze the tension between human creators and generative AI. Our theory and simulations suggest that despite challenges, a stable equilibrium between human and AI-generated content is possible. Our work contributes to understanding the
competitive dynamics in the content creation industry, offering insights into the future interplay between human creativity and technological advancements in generative AI.

Keynote: Hongning Wang

In the rapidly evolving landscape of large language models (LLMs) for medical applications, ensuring the reliability and accuracy of these models in clinical settings is paramount. Existing benchmarks often focus on fixed-format tasks like multiple-choice QA, which fail to capture the complexity of real-world clinical diagnostics. Moreover, traditional evaluation metrics and LLM-based evaluators struggle with misalignment, often providing oversimplified assessments that do not adequately reflect human judgment. To address these challenges, we introduce HDCEval, a **H**ierarchical **D**ivide-and-**C**onquer **Eval**uation framework tailored for fine-grained alignment in medical evaluation. HDCEval is built on a set of fine-grained medical evaluation guidelines developed in collaboration with professional doctors, encompassing Patient Question Relevance, Medical Knowledge Correctness, and Expression. The framework decomposes complex evaluation tasks into specialized subtasks, each evaluated by expert models trained through Attribute-Driven Token Optimization (ADTO) on a meticulously curated preference dataset. This hierarchical approach ensures that each aspect of the evaluation is handled with expert precision, leading to a significant improvement in alignment with human evaluators.

Hierarchical Divide-and-Conquer for Fine-Grained Alignment in LLM-Based Medical Evaluation

Knowledge editing aims to update outdated or incorrect knowledge in large language models (LLMs). However, current knowledge editing methods have limited scalability for lifelong editing. This study explores the fundamental reason why knowledge editing fails in lifelong editing. We begin with the closed-form solution derived from linear associative memory, which underpins state-of-the-art knowledge editing methods. We extend the solution from single editing to lifelong editing, and through rigorous mathematical derivation, identify an interference term in the final solution, suggesting that editing knowledge may impact irrelevant knowledge. Further analysis of the interference term reveals a close relationship with superposition between knowledge representations. When knowledge superposition does not exist in language models, the interference term vanishes, allowing for lossless knowledge editing. Experiments across numerous language models reveal that knowledge superposition is universal, exhibiting high kurtosis, zero mean, and heavy-tailed distributions with clear scaling laws. Ultimately, by combining theory and experiments, we demonstrate that knowledge superposition is the fundamental reason for the failure of lifelong editing. Moreover, this is the first study to investigate knowledge editing from the perspective of superposition and provides a comprehensive observation of superposition across numerous real-world language models.

Knowledge in Superposition: Unveiling the Failures of Lifelong Knowledge Editing for Large Language Models

Document Information Extraction (DIE) aims to extract structured information from Visually Rich Documents (VRDs). Previous full-training approaches have demonstrated strong performance but may struggle with generalization to unseen data. In contrast, training-free methods leverage powerful pre-trained models like Large Language Models (LLMs) to address various downstream tasks with only a few examples. Nonetheless, training-free methods for DIE encounter two primary challenges: (1) understanding the complex relationship between layout and textual elements in VRDs, and (2) providing accurate guidance to pre-trained models. To address these challenges, we propose Sample-centric In-context Learning (SAIL) for DIE. SAIL introduces a fine-grained entity-level textual similarity to facilitate in-depth text analysis by LLMs and incorporates layout similarity to enhance the analysis of layouts in VRDs. Additionally, SAIL formulates a unified In-Context Learning (ICL) prompt template for various sample-centric examples, enabling tailored prompts that deliver precise guidance to pre-trained models for each sample. Extensive experiments on FUNSD, CORD, and SROIE benchmarks with various base models (e.g., LLMs) indicate that our method outperforms training-free baselines, even closer to the full-training methods. The results show the superiority and generalization of our method.

SAIL: Sample-Centric In-Context Learning for Document Information Extraction

Graph Retrieval Augmented Generation (GRAG) is a novel paradigm that takes the naive RAG system a step further by integrating graph information, such as knowledge graph (KGs), into large-scale language models (LLMs) to mitigate hallucination. However, existing GRAG still encounter limitations: 1) simple paradigms usually fail with the complex problems due to the narrow and shallow correlations capture from KGs 2) methods of strong coupling with KGs tend to be high computation cost and time consuming if the graph is dense. In this paper, we propose the Fast Think-on-Graph (FastToG), an innovative paradigm for enabling LLMs to think "community by community" within KGs. To do this, FastToG employs  community detection for deeper correlation capture and two stages community pruning - coarse and fine pruning for faster retrieval. Furthermore, we also develop two Community-to-Text methods to convert the graph structure of communities into textual form for better understanding by LLMs. Experimental results on real-world datasets demonstrate the effectiveness of FastToG, showcasing higher accuracy, faster reasoning, and better explainability compared to baseline. Our code is open-source and available at https://xxxx.

Fast Think-on-Graph: Wider, Deeper and Faster Reasoning of Large Language Model on Knowledge Graph

Complex question answering (QA) is a challenging task in artificial intelligence research which requires reasoning based on related knowledge. The retrieval-augmented generation (RAG) based on large language models (LLMs) have become one promising solution in QA. To facilitate RAG more effectively, the LLM needs to precisely evaluate knowledge required in QA. That is, first, the LLM needs to examine its knowledge boundary (what the LLM does not know) to retrieve external knowledge as supplement. Second, the LLM needs to evaluate the utility of the retrieved knowledge (whether it helps in reasoning) for robust RAG. To this end, in this paper, we propose a novel Question Answering with Knowledge Evaluation (KEQA) framework to promote the effectiveness and efficiency of RAG in QA. First, inspired by quizzes in classroom, we propose a quiz-based method to precisely examine the knowledge state of the uninterpretable LLM for QA. We ask indicative quizzes on each required knowledge, and inspect whether the LLM can consistently answer the quiz to examine its knowledge boundary. Second, we retrieve the unknown knowledge from external source, and evaluate its utility to pick the helpful ones for reasoning. We design a reasoning-based metric to evaluate utility, and construct a demonstration set in training data for reference to guide knowledge picking in inference. We conduct extensive experiments on four widely-used QA datasets, and the results demonstrate the effectiveness of the proposed method.

Explore What LLM Does Not Know in Complex Question Answering

Continual Semantic Parsing (CSP) aims to train parsers to convert natural language questions into SQL across tasks with limited annotated examples, adapting to the real-world scenario of dynamically updated databases. Previous studies mitigate this challenge by replaying historical data or employing parameter-efficient tuning (PET), but they often violate data privacy or rely on ideal continual learning settings. To address these problems, we propose a new **L**arge Language Model (LLM)-**E**nhanced **C**ontinuous **S**emantic **P**arsing method, named **LECSP**, which alleviates forgetting while encouraging generalization, without requiring real data replay or ideal settings. Specifically, it first analyzes the commonalities and differences between tasks from the SQL syntax perspective to guide LLMs in reconstructing key memories and improving memory accuracy through a calibration strategy. Then, it uses a task-aware dual-teacher distillation framework to promote the accumulation and transfer of knowledge during sequential training.
Experimental results on two CSP benchmarks show that our method significantly outperforms existing methods, even those utilizing data replay or ideal settings. Additionally, we achieve generalization performance beyond the upper limits, better adapting to unseen tasks.

Filling Memory Gaps: Enhancing Continual Semantic Parsing via SQL Syntax Variance-Guided LLMs Without Real Data Replay

Large Language Models (LLMs) excel in linguistic tasks but struggle with mathematical reasoning, particularly in non- English languages like Hindi. This research aims to en- hance the mathematical reasoning skills of smaller, resource- efficient open-source LLMs in both Hindi and English. We evaluate models like OpenHathi 7B, LLaMA-2 7B, Wizard- Math 7B, Mistral 7B, LLeMMa 7B, MAmmoTH 7B, Gemini Pro, and GPT-4 using zero-shot, few-shot chain-of-thought (CoT) methods, and supervised fine-tuning. Our approach in- corporates curriculum learning, progressively training mod- els on increasingly difficult problems, a novel Decompo- sition Strategy to simplify complex arithmetic operations, and a Structured Solution Design that divides solutions into phases. Our experiments result in notable performance en- hancements. WizardMath 7B exceeds Gemini’s accuracy on English datasets by +6% and matches Gemini’s performance on Hindi datasets. Adopting a bilingual approach that com- bines English and Hindi samples achieves results comparable to individual language models, demonstrating the capability to learn mathematical reasoning in both languages. This re- search highlights the potential for improving mathematical reasoning in open-source LLMs.

Multilingual Mathematical Reasoning: Advancing Open-Source LLMs in Hindi and English

Recent advancements in large language models (LLMs) have broadened their application scope but revealed challenges in balancing capabilities across general knowledge, coding, and mathematics. To address this, we introduce a Collaborative and Semantic Experts (CoE) approach for supervised fine-tuning (SFT), which employs a two-phase training strategy. Initially, expert training fine-tunes the feed-forward network on specialized datasets, developing distinct experts in targeted domains. Subsequently, expert leveraging synthesizes these trained experts into a structured model with semantic guidance to activate specific experts, enhancing performance and interpretability. Evaluations on comprehensive benchmarks across MMLU, HumanEval, GSM8K, MT-Bench, and AlpacaEval confirm CoE's efficacy, demonstrating improved performance and expert collaboration in diverse tasks, significantly outperforming traditional SFT methods. We commit to releasing all training data and models, anticipating that this will contribute to the advancement of open-source LLMs.

Fine-Tuning Language Models with Collaborative and Semantic Experts

Emotion Recognition in Conversations (ERC) refers to the task of automatically identifying the emotion of each utterance in conversations. Unfortunately, because the emotion of an utterance is influenced by the contextual information of the conversation, to label each utterance in ERC, the annotators are required to repeatedly read the whole conversation, making the fine-grained labeling process very costly. In this work, we suggest an efficient coarse-grained labeling pattern for ERC, where it asks annotators to assign a set of emotions for each conversation. The empirical results of our preliminary evaluations indicate the accuracy and efficiency of this conversation-level labeling pattern. Accordingly, we suggest that inducing ERC predictors with conversation-level emotion sets in a weakly-supervised learning manner can be a potential candidate for ERC, and we formally refer to this new weakly-supervised learning task as Dataless ERC (DERC). To validate its potential, we propose a simple-yet-flexible DERC framework, namely DERC with Progressive Learning (DERC-PL). We jointly update pseudo-utterance-level emotions and the ERC predictor in a self-training manner, and in particular, progressively update the ERC predictor from training subsets with lower noise densities to the ones with higher noise densities. We specify several versions of DERC-PL by applying various off-the-shelf ERC methods. Empirical results indicate that DERC-PL can be on par with existing weakly-supervised learning baselines and supervised learning ERC methods.

Utterance-level Emotion Recognition in Conversation with Conversation-level Supervision

Instruction Fine-Tuning (IFT) significantly enhances the zero-shot capabilities of pretrained Large Language Models (LLMs). While coding data is known to boost LLM reasoning abilities during pretraining, its role in activating internal reasoning capacities during IFT remains understudied. This paper investigates a key question: How does coding data impact LLMs' reasoning capacities during IFT stage? To explore this, we thoroughly examine the impact of coding data across different coding data proportions, model families, sizes, and reasoning domains, from various perspectives. Specifically, we create three IFT datasets with increasing coding data proportions, fine-tune six LLM backbones across different families and scales on these datasets, evaluate the tuned models' performance across twelve tasks in three reasoning domains, and analyze the outcomes from three broad-to-granular perspectives: overall, domain-level, and task-specific.
Our holistic analysis provides valuable insights into each perspective. First, coding data tuning enhances the overall reasoning capabilities of LLMs across different model families and scales. Moreover, while the impact of coding data varies by domain, it shows consistent trends within each domain across different model families and scales. Additionally, coding data generally provides comparable task-specific benefits across model families, with optimal proportions in IFT datasets being task-dependent.

Unveiling the Impact of Coding Data Instruction Fine-Tuning on Large Language Models Reasoning

Despite the rapid progress that existing automated feedback methods have made in correcting the output of large language models (LLMs), these methods cannot be well applied to the relation extraction (RE) task due to their designated feedback objectives and correction manner. To address this problem, we propose a novel automated feedback framework for RE, which presents a rationale supervisor to verify the rationale and provide re-selected demonstrations as feedback to correct the initial prediction. Specifically, we first design a causal intervention and observation method for to collect biased/unbiased rationales for contrastive training the rationale supervisor. Then, we present a verification-feedback-correction procedure to iteratively enhance LLMs' capability of handling the RE task. Extensive experiments prove that our proposed framework significantly outperforms existing methods.

Enhancing Relation Extraction via Supervised Rationale Verification and Feedback

Large Language Models (LLMs) have significantly advanced natural language processing (NLP), providing versatile capabilities across various applications. However, their application to complex, domain-specific tasks, such as cyber-security, often faces substantial challenges. In this study, we introduce SecKnowledge and CyberPal.AI to address these challenges and train security-expert LLMs.
SecKnowledge is a domain-knowledge-driven cyber-security instruction dataset, meticulously designed using years of accumulated expert knowledge in the domain through a multi-phase generation process. CyberPal.AI refers to a family of LLMs fine-tuned using SecKnowledge, aimed at building security-specialized LLMs capable of answering and following complex security-related instructions. 
Additionally, we introduce SecKnowledge-Eval, a comprehensive and diverse cyber-security evaluation benchmark, composed of an extensive set of cyber-security tasks we specifically developed to assess LLMs in the field of cyber-security, along with other publicly available security benchmarks.
Our results show a significant average improvement of up to 24% over the baseline models, underscoring the benefits of our expert-driven instruction dataset generation process.
These findings contribute to the advancement of AI-based cyber-security applications, paving the way for security-expert LLMs that can enhance threat-hunting and investigation processes.

CyberPal.AI: Empowering LLMs with Expert-Driven Cybersecurity Instructions

Large Language Models (LLMs) have shown great promise in vulnerability identification. As C/C++ comprise half of the Open-Source Software (OSS) vulnerabilities over the past decade and updates in OSS mainly occur through commits, enhancing LLMs' ability to identify C/C++ Vulnerability-Contributing Commits (VCCs) is essential. However, current studies primarily focus on further pre-training LLMs on massive code datasets, which is resource-intensive and poses efficiency challenges.
In this paper, we enhance the ability of BERT-based LLMs to identify C/C++ VCCs in a lightweight manner. We propose CodeLinguaNexus (CLNX) as a bridge facilitating communication between C/C++ programs and LLMs. Based on commits, CLNX efficiently converts the source code into a more natural representation while preserving key details. Specifically, CLNX first applies structure-level naturalization to decompose complex programs, followed by token-level naturalization to interpret complex symbols. We evaluate CLNX on public datasets of 25,872 C/C++ functions with their commits. The results demonstrate that CLNX substantially improves the ability of LLMs to detect C/C++ VCCs. Moreover, CLNX-equipped CodeBERT achieves new state-of-the-art and identifies 38 OSS vulnerabilities in the real world.

CLNX: Bridging Code and Natural Language for C/C++ Vulnerability-Contributing Commits Identification

Large Language Models (LLMs) are known to hallucinate facts and make non-factual statements which can undermine trust in their output.
The essence of hallucination lies in the absence of metacognition in LLMs, namely the understanding of their own cognitive processes. However, there has been limited research on quantitatively measuring metacognition within LLMs. Drawing inspiration from cognitive psychology theories, we first quantify the metacognitive ability of LLMs as their ability to evaluate the correctness of responses through confidence. Subsequently, we introduce a general framework called DMC designed to decouple metacognitive ability and cognitive ability. This framework tackles the challenge of noisy quantification caused by the coupling of metacognition and cognition in current research, such as calibration-based metrics. Specifically, the DMC framework comprises two key steps. Initially, the framework tasks the LLM with failure prediction, aiming to evaluate the model's performance in predicting failures, a performance jointly determined by both cognitive and metacognitive abilities of the LLM. Following this, the framework disentangles metacognitive ability and cognitive ability based on the failure prediction performance, providing a quantification of the LLM's metacognitive ability independent of cognitive influences. Experiments conducted on eight datasets across five domains reveal that (1) Our proposed DMC framework effectively separates the metacognition and cognition of LLMs; (2) Various confidence elicitation methods impact the quantification of metacognitve ability differently; (3) Stronger metacognitive ability are exhibited by LLMs with better overall performance; (4) Enhancing metacognition holds promise for alleviating hallucination issues.

Decoupling Metacognition from Cognition: A Framework for Quantifying Metacognitive Ability in LLMs

With the outstanding capabilities of Large Language Models (LLMs), 
solving math word problems (MWP) has greatly progressed, achieving higher performance on several benchmark datasets. 
However, it is more challenging to solve plane geometry problems (PGPs) due to the necessity of understanding, reasoning and computation on two modality data including both geometry diagrams and textual questions, where Multi-Modal Large Language Models (MLLMs) have not been extensively explored.
Previous works simply regarded a plane geometry problem as multi-modal QA task, which ignored the importance of explicit parsing geometric elements from problems. 
To tackle this limitation, we propose to solve plane Geometry problems by Neural-Symbolic reasoning with MLLMs (GNS). 
We first leverage an MLLM to understand PGPs through knowledge prediction and symbolic parsing, next perform mathematical reasoning to obtain solutions, last adopt a symbolic solver to compute answers. 
Correspondingly, we introduce the largest PGPs dataset GNS-260K with multiple annotations including symbolic parsing, understanding, reasoning and computation. 
In experiments, our Phi3-Vision-based MLLM wins the first place on the PGPs solving task of MathVista benchmark, outperforming GPT-4o, Gemini Ultra and other much larger MLLMs.
While LLaVA-13B-based MLLM markedly exceeded other close-source and open-source MLLMs on the MathVerse benchmark and also achieved the new SOTA on GeoQA dataset.
We will make the proposed models, prompts and dataset publicly available.

GNS: Solving Plane Geometry Problems by Neural-Symbolic Reasoning with Multi-Modal LLMs

Few-shot Named Entity Recognition (NER) spotlights the tag of novel entity types in data-limited scenarios or lower-resource settings. Advances with Pre-trained Language Models (PLMs), including BERT, GPT, and their variants, have driven tremendous strategies to leverage context-dependent representations and exploit predefined relational cues, yielding significant gains in witnessing unseen entities. Nevertheless, a fundamental issue exists in prior efforts regarding their susceptibility to adversarial attacks in the intricate semantic environment. This vulnerability undermines the robustness of semantic representations, exacerbating the challenge of accurate entity identification, especially when transitioning across domains. To this end, we propose an $A$dversity-aware $A$ugment $L$earning ($AAL$) solution for the few-shot NER task, dedicated to retrieving and reinforcing entity prototypes resilient to adversarial inference, thereby enhancing cross-domain semantic coherence. In particular, AAL employs a two-stage paradigm consisting of training and fine-tuning. The process initiates with augmentation learning by leveraging two kinds of prompt learning schemes, then identifies prototypes under the guidance of a variational manner. Furthermore, we devise a domain-oriented prototype refinement to optimize prototype learning under conditions of uncertainty attack, facilitating the effective transfer of common knowledge from source to target domains. The experimental results, encompassing the few-shot NER datasets under both certainty and uncertainty conditions, affirm the superiority of the proposed AAL over several representative baselines, particularly its capability against adversarial attacks.

Adversity-aware Few-shot Named Entity Recognition via Augmentation Learning

Natural Language Processing 15

We are pleased to announce the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-25), which will be held in Philadelphia, Pennsylvania at the Pennsylvania Convention Center from February 25 to March 4, 2025.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.

### [Invited Speakers](https://aaai.org/conference/aaai/aaai-25/aaai-25-invited-speakers/)

Register [here](https://aaai.org/conference/aaai/aaai-25/registration/)

AAAI 2025

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-25 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.



Underline digital video library

Browse Experimental Biology

Browse Computational Linguistics Keynote and Invited Talks

Browse Underline Originals

No results!

Browse Natural Language Processing

Browse IEEE International Reliability Physics Symposium (IRPS) 2024 Keynotes

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES