United States

We introduce Functional Group-Aware Representations for Small Molecules (FARM), a novel foundation model designed to bridge the gap between SMILES, natural language, and molecular graphs. The key innovation of FARM lies in its functional group-aware tokenization, which directly incorporates functional group information into SMILES, enriching SMILES with detailed chemical context. For example, instead of using ``O&quot; to represent all oxygen atoms, we use specific tokens like &quot;O_ketone&quot; and &quot;O_hydroxyl&quot; to differentiate oxygen atoms belonging to distinct functional groups. This tokenization expands the chemical lexicon, effectively bridging the gap between SMILES and natural language in terms of vocabulary size, ultimately enhancing the model’s ability to predict molecular properties. FARM also represents molecules from two perspectives: by (1) using masked language modeling to capture atom-level features and (2) employing graph neural networks to encode the whole molecule topology. FARM leverages contrastive learning to aligns these two views of representations into a unified molecular embedding. We rigorously evaluate FARM on the MoleculeNet dataset, where it achieves state-of-the-art performance on 11 out of 13 tasks. These results highlight FARM’s potential to improve molecular representation learning and demonstrate its strong transfer learning capabilities, paving the way for promising applications in drug discovery and pharmaceutical research.

NAACL 2025

FARM: Functional Group-Aware Representations for Small Molecules

functional group-aware tokenization

molecular representation learning

contrastive learning

The OWSD General Assembly and international conference will be held entirely online for the first time in OWSD's 28-year history, allowing as many of OWSD's 7600+ members to participate as possible.

The conference theme is 'Women, science and development', exploring three areas of relevance: 1) the importance of basic sciences for development; 2) the impact of applied sciences on development and 3) including sex and gender as variables in scientific research.


**Keynote speakers include:** 
* Fabiola Gianotti, Director-General of CERN; 
* Bina Agarwal, an internationally celebrated economist specializing in the political economy of gender, poverty and inequality; and 
* Angela Saini, award-winning author of Inferior: How Science Got Women Wrong and the New Research That’s Rewriting the Story, Superior: The Return of Race Science, and other books. 

OWSD members will present their own research in the three themes during poster sessions throughout the conference. In addition, members will also present their own scientific and personal responses to the COVID-19 pandemic. All members will also have the opportunity to take part in a workshop on increasing their visibility and impact as a scientist, with a session on using podcasts for storytelling by Joy Owango and Adelle Onyango of the Training Centre in Communication (TCC Africa), followed by a session on filmmaking for impact by Nicole Leghissa, OWSD video consultant.

The conference will include a handover of the previous OWSD Executive Board to the new members, elected in June-July 2021. On the last day of the conference (19 November), the OWSD Italy Country Network will be launched. **This will be an in-person event in Trieste.** Italian government representatives and donors are invited to attend. The Trieste event will be coordinated with similar in person events being held in OWSD countries throughout the developing world and streamed online. At least 2,000 OWSD members, women scientists from developing countries around the world, are expected to participate online in the General Assembly and conference.

 

**ABOUT THE PROGRAMME**

In order to approach the complex and wide-ranging challenges of the 2030 Agenda for Sustainable Development, societies around the world must access and harness all the talent and experience available to them. The concerns and perceptions of the greatest number of all possible actors and recipients must be considered in order that the most relevant issues and effective solutions are found. Women are considerably underrepresented as researchers in STEM subjects around the globe; according to UNESCO Institute of Statistics (UIS) data, in 2019 the global average percentage of women researchers was still below 30%. Yet women typically have first hand specialist knowledge of the day-to-day challenges diverse individuals face within communities, and can be key informers in the design and development of their solutions. Moreover, since women are most often the providers of informal community support, if they are not fully informed—and if solutions are developed without their consultation—it may be the case that products, services and solutions are not relevant or appropriate. Technological advances may mean improved machines on large farms, for example, potentially increasing food production, but since the key points of food insecurity in the world are on small-scale farms managed by women throughout the Global South, such capital investments will have little impact in the areas where food production is most critical—and where women’s knowledgeable implementation of appropriate innovative technologies could make the most difference.

 
 


**Keynote Speakers:**
* Fabiola Gianotti, Director-General of CERN; 
* Bina Agarwal, an internationally celebrated economist specializing in the political economy of gender, poverty and inequality; and 
* Angela Saini, award-winning author of Inferior: How Science Got Women Wrong and the New Research That’s Rewriting the Story, Superior: The Return of Race Science, and other books. 

**Eminent Guests:**
* Elena Bonetti, Italian Minister for Family and Equal Opportunities
* Maria Cristina Messa,	Italian Minister of Universities and Research
* Shamila Nair-Bedouelle,	Assistant Director General for Natural Sciences, UNESCO
* Maria Chiara Carrozza,	President, National Research Council of Italy
* Kerstin Jonsson Cisse,	Head of Unit for Globally Sustainable Economic Development, Sida - Sweden
* Naser Faruqui,	Director of Education and Science at IDRC - Canada
* Ylann Schemm,	Director, Elsevier Foundation
* Alberto Quadrio Curzio,	President Emeritus of the Italian Academy of Sciences and Humanities 
* Mohamed Hassan,	President, TWAS the World Academy of Sciences

**Chairs, panelists, interviewers and other speakers:**
Jennifer Thomson, Tonya Blowers, Rana Dajani, Shobhana Narasimhan, Joy Owango, Elizabeth Pollitzer, Kausila Timsina, Elizabeth Bandason, Susana Arrechea, Reinabelle Reyes, Dorine Odongo, Elizabeth Pollitzer, Adelle Onyango, Nicole Leghissa, Jana Rodriguez Hertz, Maryse Nkoua Ngavouka, Kate Shaw, Chao Mbogo, Prativa Pandey, Edna Patricia Matta-Camacho, Amira Shaheen, Wezi Nyambili, Ana Lucía Martinez Abarca, Mashura Shammi, Therese Umuhoza, Dionica Gamboa, Marian Nkansah, Babalola Olubukola Oluranti, Esi Awuah, Fortunate Farirai, Atya Kapley, Hasin Anupama Azhari, Huda Basaleem, Nashwa Eassa, Shymaa Enany, Kleinsy Bonilla, Patricia Castillo-Briceno, Fang Xin

Please register for the conference to be able to enter the event page on Underline.io. If you have already registered, please login to Underline.io to be able to see full conference content. In case you do not have an Underline account, please create one. Thank you

Please login to your Underline account

OWSD 6th GENERAL ASSEMBLY AND INTERNATIONAL CONFERENCE

The conference theme is 'Women, science and development', exploring three areas of relevance: the importance of basic sciences for development; the impact of applied sciences on development, and including sex and gender as variables in scientific research.

We introduce Functional Group-Aware Representations for Small Molecules (FARM), a novel foundation model designed to bridge the gap between SMILES, natural language, and molecular graphs. The key innovation of FARM lies in its functional group-aware tokenization, which directly incorporates functional group information into SMILES, enriching SMILES with detailed chemical context. For example, instead of using ``O" to represent all oxygen atoms, we use specific tokens like "O_ketone" and "O_hydroxyl" to differentiate oxygen atoms belonging to distinct functional groups. This tokenization expands the chemical lexicon, effectively bridging the gap between SMILES and natural language in terms of vocabulary size, ultimately enhancing the model’s ability to predict molecular properties. FARM also represents molecules from two perspectives: by (1) using masked language modeling to capture atom-level features and (2) employing graph neural networks to encode the whole molecule topology. FARM leverages contrastive learning to aligns these two views of representations into a unified molecular embedding. We rigorously evaluate FARM on the MoleculeNet dataset, where it achieves state-of-the-art performance on 11 out of 13 tasks. These results highlight FARM’s potential to improve molecular representation learning and demonstrate its strong transfer learning capabilities, paving the way for promising applications in drug discovery and pharmaceutical research.

workshop paper

### Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Welcome to the 2025 meeting of the Nations of the Americas Chapter of the Association for Computational Linguistics! I am proud to help organize the first NAACL conference to carry the new name of our organization, one that emphasizes inclusion for all of the Americas. I am also pleased to welcome you to Albuquerque, New Mexico, a state whose unique blend of cultural influences will make for an excellent backdrop for NAACL 2025, especially with this year’s special theme on NLP in a Multicultural World. 
**[Continue reading...](https://drive.google.com/file/d/1jX-qGhqVSZZCIrAnJaz798pflu5Irrdn/view?usp=sharing)**

*- Colin Cherry, Google, NAACL 2025 General Chair* 

[![](https://assets.underline.io/markdown_image/1/image/b087f8a4dc5816d6e1a6514e59c59ac3.png)](https://drive.google.com/file/d/1T96GzPqObXrMln2BMByCSXTSizjTg69P/view?usp=sharing)

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.

To access NAACL 2025 event page you are required to register. Please follow [**this link**](https://2025.naacl.org/registration/registration/) to register. Access will depend on your registration type.

Please register!

Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Automatic speech recognition (ASR) systems face many sources of extra-linguistic noise that ideally should not affect their transcription of a speech signal. This includes speaker factors, recording quality, or even ``adversarial noise'' designed to disrupt performance. We experimented with the dialect composition of ASR fine-tuning data to study whether balancing the relative proportion of dialects had an impact on models' robustness to two such sources of variation: dialect variation and adversarial perturbations. We fine-tuned XLSR-53 for Catalan ASR using four different dialect compositions. These were defined as 100%, 80%, 50%, and 20% Central Catalan, with the remaining portions split evenly between four other Catalan dialects. While increasing the relative proportion of dialect variants improved models' dialect robustness, this did not have a meaningful impact on adversarial robustness. These findings suggest that while improvements to ASR can be made by diversifying the training data, such changes do not sufficiently counteract adversarial attacks, leaving the technology open to security threats.

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

This paper examines the linguistic distinction between syntax and morphology, focusing on noun-noun compounds in three West Germanic languages (English, Dutch, and German). Previous studies using the Parallel Bible Corpus have found a trade-off between word order (syntax) and word structure (morphology), with languages optimizing information conveyance through these systems. Our research question is whether manipulating English noun-noun compounds to resemble Dutch and German constructions can reproduce the observed distance between these languages in the order-structure plane. We extend a word-pasting procedure to merge increasingly common noun-noun pairs in English Bible translations. After each merge, we estimate the information contained in word order and word structure using entropy calculations. Our results show that pasting noun-noun pairs reduces the difference between English and the other languages, suggesting that orthographic conventions defining word boundaries play a role in this distinction. However, the effect is not pronounced, and results for two of the six books considered are inconclusive.

West Germanic noun-noun compounds and the morphology-syntax trade-off

Partly automated creation of interlinear glossed text (IGT) has the potential to assist in linguistic documentation. We argue that LLMs can make this process more accessible to linguists because of their capacity to follow natural-language instructions. We investigate the effectiveness of a retrieval-based LLM prompting approach to glossing, applied to the seven languages from the SIGMORPHON 2023 shared task. Our system beats the BERT-based shared task baseline for every language in the morpheme-level score category, and we show that a simple 3-best oracle has higher word-level scores than the challenge winner (a tuned sequence model) in five languages. In a case study on Tsez, we ask the LLM to automatically create and follow linguistic instructions, reducing errors on a confusing grammatical feature. Our results thus demonstrate the potential contributions which LLMs can make in interactive systems for glossing, both in making suggestions to human annotators and following directions.

Prompt and circumstance: A word-by-word LLM prompting approach to interlinear glossing for low-resource languages

Large Language Models (LLMs) have achieved remarkable performance across a wide variety of natural language tasks. However, they have been shown to suffer from a critical limitation pertinent to 'hallucination' in their output. Recent research has focused on investigating and addressing this problem for a variety of tasks such as biography generation, question answering, abstractive summarization, and dialogue generation. However, the crucial aspect pertaining to 'negation' has remained considerably underexplored. Negation is important because it adds depth and nuance to the understanding of language and is also crucial for logical reasoning and inference. In this work, we address the above limitation and particularly focus on studying the impact of negation in LLM hallucinations. Specifically, we study four tasks with negation: 'false premise completion', 'constrained fact generation', 'multiple choice question answering', and 'fact generation'. We show that open-source state-of-the-art LLMs such as LLaMA-2-chat, Vicuna, and Orca-2 hallucinate considerably on all these tasks involving negation which underlines a critical shortcoming of these models. Addressing this problem, we further study numerous strategies to mitigate these hallucinations and demonstrate their impact.

Investigating and Addressing Hallucinations of LLMs in Tasks Involving Negation

In this paper, we study whether language models are affected by learned gender stereotypes during the comprehension of stories. Specifically, we investigate how models respond to gender stereotype perturbations through counterfactual data augmentation. Focusing on Question Answering (QA) tasks in fairytales, we modify the FairytaleQA dataset by swapping gendered character information and introducing counterfactual gender stereotypes during training. This allows us to assess model robustness and examine whether learned biases influence story comprehension. Our results show that models exhibit slight performance drops when faced with gender perturbations in the test set, indicating sensitivity to learned stereotypes. However, when fine-tuned on counterfactual training data, models become more robust to anti-stereotypical narratives. Additionally, we conduct a case study demonstrating how incorporating counterfactual anti-stereotype examples can improve inclusivity in downstream applications.

Will the Prince Get True Love's Kiss? On the Model Sensitivity to Gender Perturbation over Fairytale Texts

Language models (LMs) are vulnerable to exploitation for adversarial misuse. Training LMs for safety alignment is extensive, making it hard to respond to fast-developing attacks immediately, such as jailbreaks. We propose self-refine with formatting that achieves outstanding safety even in non-safety-aligned LMs and evaluate our method alongside several defense baselines, demonstrating that it is the safest training-free method against jailbreak attacks. Additionally, we proposed a formatting method that improves the efficiency of the self-refine process while reducing attack success rates in fewer iterations. We observed that non-safety-aligned LMs outperform safety-aligned LMs in safety tasks by giving more helpful and safe responses. In conclusion, our findings can achieve less safety risk with fewer computational costs, allowing non-safety LM to be efficiently utilized in real-world service.

Break the Breakout: Reinventing LM Defense Against Jailbreak Attacks with Self-Refine

Food systems are responsible for a third of human-caused greenhouse gas emissions. We investigate what Large Language Models (LLMs) can contribute to reducing the environmental impacts of food production. We define a typology of design and prediction tasks based on the sustainable food literature and collaboration with domain experts, and evaluate six LLMs on four tasks in our typology. For example, for a sustainable protein design task, food science experts estimated that collaboration with an LLM can reduce time spent by 45% on average, compared to 22% for collaboration with another expert human food scientist. However, for a sustainable menu design task, LLMs produce suboptimal solutions when instructed to consider both human satisfaction and climate impacts. We propose a general framework for integrating LLMs with combinatorial optimization to improve reasoning capabilities. Our approach decreases emissions of food choices by 79% in a hypothetical restaurant while maintaining participants' satisfaction. Our results demonstrate LLMs' potential, supported by optimization techniques, to accelerate sustainable food development and adoption.

What Can Large Language Models Do for Sustainable Food?

Effective generation of novel hypotheses is instrumental to scientific progress. So far, researchers have been the main powerhouse behind hypothesis generation by painstaking data analysis and thinking (also known as the Eureka moment). In this paper, we examine the potential of large language models (LLMs) to generate hypotheses. We focus on hypothesis generation based on data (i.e., labeled examples). To enable LLMs to handle arbitrarily long contexts, we generate initial hypotheses from a small number of examples and then update them iteratively to improve the quality of hypotheses. Inspired by multi-armed bandits, we design a reward function to inform the exploitation-exploration tradeoff in the update process. Our algorithm is able to generate hypotheses that enable much better predictive performance than few-shot prompting in classification tasks, improving accuracy by 31.7% on a synthetic dataset and by 13.9%, 3.3% and, 24.9% on three real-world datasets. We also outperform supervised learning by 12.8% and 11.2% on two challenging real-world datasets. Furthermore, we find that the generated hypotheses not only corroborate human-verified theories but also uncover new insights for the tasks.

Hypothesis Generation with Large Language Models

Citation practices are crucial in shaping the structure of scientific knowledge, yet they are often influenced by contemporary norms and biases. The emergence of Large Language Models (LLMs) introduces a new dynamic to these practices. Interestingly, the characteristics and potential biases of references recommended by LLMs that entirely rely on their parametric knowledge, and not on search or retrieval-augmented generation, remain unexplored. Here, we analyze these characteristics in an experiment using a dataset from AAAI, NeurIPS, ICML, and ICLR, published after GPT-4's knowledge cut-off date. In our experiment, LLMs are tasked with suggesting scholarly references for the anonymized in-text citations within these papers. Our findings reveal a remarkable similarity between human and LLM citation patterns, but with a more pronounced high citation bias, which persists even after controlling for publication year, title length, number of authors, and venue. The results hold for both GPT-4, and the more capable models GPT-4o and Claude 3.5 where the papers are part of the training data. Additionally, we observe a large consistency between the characteristics of LLM's existing and non-existent generated references, indicating the model's internalization of citation patterns. By analyzing citation graphs, we show that the references recommended are embedded in the relevant citation context, suggesting an even deeper conceptual internalization of the citation networks. While LLMs can aid in citation generation, they may also amplify existing biases, such as the Matthew effect, and introduce new ones, potentially skewing scientific knowledge dissemination.

Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias

Automated scientific discovery promises to accelerate progress across scientific domains. However, developing and evaluating an AI agent's capacity for end-to-end scientific reasoning is challenging as running real-world experiments is often prohibitively expensive or infeasible. In this work we introduce DISCOVERYWORLD, the first virtual environment for developing and benchmarking an agent's ability to perform complete cycles of novel scientific discovery. DISCOVERYWORLD contains a variety of different challenges, covering topics as diverse as radioisotope dating, rocket science, and proteomics, to encourage development of general discovery skills rather than task-specific solutions. DISCOVERYWORLD itself is an inexpensive, simulated, text-based environment (with optional 2D visual overlay). It includes 120 different challenge tasks, spanning eight topics each with three levels of difficulty and several parametric variations. Each task requires an agent to form hypotheses, design and run experiments, analyze results, and act on conclusions. DISCOVERYWORLD further provides three automatic metrics for evaluating performance, based on (a) task completion, (b) task-relevant actions taken, and (c) the discovered explanatory knowledge. We find that strong baseline agents, that perform well in prior published environments, struggle on most DISCOVERYWORLD tasks, suggesting that DISCOVERYWORLD captures some of the novel challenges of discovery, and thus that DISCOVERYWORLD may help accelerate near-term development and assessment of scientific discovery competency in agents. Code available at: http://www.github.com/allenai/discoveryworld

Downloads

Next from NAACL 2025

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

Similar lecture

Rethinking the way we farm

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from NAACL 2025

The Impact of Dialect Variation on Robust Automatic Speech Recognition for Catalan

Similar lecture

Rethinking the way we farm

Downloads