United States

Large language models (LLMs) have achieved remarkable success but still tend to generate factually erroneous responses, a phenomenon known as hallucination. A recent trend is to use preference learning to fine-tune models to align with factuality. However, existing work primarily evaluates fine-tuned models on in-domain (ID) datasets and the factuality on out-of-domain (OOD) datasets remains underexplored. In this paper, we conduct a comprehensive evaluation of the factuality of different models tuned by various preference learning algorithms and demonstrate that their performance on OOD datasets either increases minimally or decreases. Subsequently, we reveal that the main cause of model&#39;s failure to uphold factuality under a distribution shift is **under-alignment**, rather than **over-alignment**, by analyzing the token distribution shift of the models before and after tuning. Finally, we propose **APEFT** (**A**tomic **P**reference **E**nhanced **F**actuality **T**uning), a framework that enhances model&#39;s awareness of factuality at the granularity of individual facts. Extensive experiments demonstrate that APEFT improves model performance by an average of on both ID and OOD datasets, which is highly effective.

NAACL 2025

Beyond Under-Alignment: Atomic Preference Enhanced Factuality Tuning for Large Language Models

factuality tuning

llms

hallucination

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

Large language models (LLMs) have achieved remarkable success but still tend to generate factually erroneous responses, a phenomenon known as hallucination. A recent trend is to use preference learning to fine-tune models to align with factuality. However, existing work primarily evaluates fine-tuned models on in-domain (ID) datasets and the factuality on out-of-domain (OOD) datasets remains underexplored. In this paper, we conduct a comprehensive evaluation of the factuality of different models tuned by various preference learning algorithms and demonstrate that their performance on OOD datasets either increases minimally or decreases. Subsequently, we reveal that the main cause of model's failure to uphold factuality under a distribution shift is **under-alignment**, rather than **over-alignment**, by analyzing the token distribution shift of the models before and after tuning. Finally, we propose **APEFT** (**A**tomic **P**reference **E**nhanced **F**actuality **T**uning), a framework that enhances model's awareness of factuality at the granularity of individual facts. Extensive experiments demonstrate that APEFT improves model performance by an average of on both ID and OOD datasets, which is highly effective.

poster

### 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

This paper investigates the automation of qualitative data analysis, focusing on inductive coding using large language models (LLMs). Unlike traditional approaches that rely on deductive methods with predefined labels, this research investigates the inductive process where labels emerge from the data. The study evaluates the performance of six open-source LLMs compared to human experts. As part of the evaluation, experts rated the perceived difficulty of the quotes they coded. The results reveal a peculiar dichotomy: human coders consistently perform well when labeling complex sentences but struggle with simpler ones, while LLMs exhibit the opposite trend. Additionally, the study explores systematic deviations in both human and LLM-generated labels by comparing them to the golden standard from the test set. While human annotations may sometimes differ from the golden standard, they are often rated more favorably by other humans. In contrast, some LLMs demonstrate closer alignment with the true labels but receive lower evaluations from experts.

Text Annotation via Inductive Coding: Comparing Human Experts to LLMs in Qualitative Data Analysis

Large Language Models (LLMs) have significantly advanced legal intelligence, but the scarcity of scenario data impedes the progress toward interactive legal scenarios. This paper introduces a Multi-agent Legal Simulation Driver (MASER) to scalably generate synthetic data by simulating interactive legal scenarios. Leveraging real-legal case sources, MASER ensures the consistency of legal attributes between participants and introduces a supervisory mechanism to align participants' characters and behaviors as well as addressing distractions. A Multi-stage Interactive Legal Evaluation (MILE) benchmark is further constructed to evaluate LLMs’ performance in dynamic legal scenarios. Extensive experiments confirm the effectiveness of our framework.

Multi-Agent Simulator Drives Language Models for Legal Intensive Interaction

This paper delves into sequence-level knowledge distillation (KD) of multilingual pre-trained translation models. We posit that, beyond the approximated mode obtained via beam search, the whole output distribution of the teacher contains valuable insights for students. We explore the potential of n-best lists from beam search to guide student's learning and then investigate alternative decoding methods to address observed issues like low variability and under-representation of infrequent tokens. Our research in data-limited scenarios reveals that although sampling methods can slightly compromise the translation quality of the teacher output compared to beam search based methods, they enrich the generated corpora with increased variability and lexical richness, ultimately enhancing student model performance and reducing the gender bias amplification commonly associated with KD.

Beyond the Mode: Sequence-Level Distillation of Multilingual Translation Models for Low-Resource Language Pairs

In this study, we introduce **BanNERD**, the most extensive human-annotated and validated **B**angla **N**amed **E**ntity **R**ecognition **D**ataset to date, comprising over 85,000 sentences. **BanNERD** is curated from a diverse array of sources, spanning over 29 domains, thereby offering a comprehensive range of generalized contexts. To ensure the dataset's quality, expert linguists developed a detailed annotation guideline tailored to the Bangla language. All annotations underwent rigorous validation by a team of validators, with final labels being determined via majority voting, thereby ensuring the highest annotation quality and a high IAA score of 0.88. In a cross-dataset evaluation, models trained on BanNERD consistently outperformed those trained on four existing Bangla NER datasets. Additionally, we propose a method named **BanNERCEM** (Bangla NER context-ensemble Method) which outperforms existing approaches on Bangla NER datasets and performs competitively on English datasets using lightweight Bangla pretrained LLMs. Our approach passes each context separately to the model instead of previous concatenation-based approaches achieving the highest average macro F1 score of 81.85% across 10 NER classes, outperforming previous approaches and ensuring better context utilization. We are making the code and datasets publicly available at https://github.com/eblict-gigatech/BanNERD in order to contribute to the further advancement of Bangla NLP.

BanNERD: A Benchmark Dataset and Context-Driven Approach for Bangla Named Entity Recognition

Healthcare professionals are increasingly including Language Models (LMs) in clinical practice. However, LMs have been shown to exhibit and amplify stereotypical biases that can cause life-threatening harm in a medical context. This study aims to evaluate gender biases in automatically generated clinical cases in French, on ten disorders. Using seven LMs fine-tuned for clinical case generation and an automatic linguistic gender detection tool, we measure the associations between disorders and gender. We unveil that LMs over-generate cases describing male patients, creating synthetic corpora that are not consistent with documented prevalence for these disorders. For instance, when prompts do not specify a gender, LMs generate eight times more clinical cases describing male (vs. female patients) for heart attack. We discuss the ideal synthetic clinical case corpus and establish that explicitly mentioning demographic information in generation instructions appears to be the fairest strategy. In conclusion, we argue that the presence of gender biases in synthetic text raises concerns about LM-induced harm, especially for women and transgender people.

Women do not have heart attacks!'' Gender Biases in Automatically Generated Clinical Cases in French

Hate speech detection deals with many language variants, slang, slurs, expression modalities, and cultural nuances. This outlines the importance of working with specific corpora, when addressing hate speech within the scope of Natural Language Processing, recently revolutionized by the irruption of Large Language Models. This work presents a brief analysis of the performance of large language models in the detection of Hate Speech for Rioplatense Spanish. We performed classification experiments leveraging chain-of-thought reasoning with ChatGPT 3.5, Mixtral, and Aya, comparing their results with those of a state-of-the-art BERT classifier. These experiments outline that, even if large language models show a lower precision compared to the fine-tuned BERT classifier and, in some cases, they find hard-to-get slurs or colloquialisms, they still are sensitive to highly nuanced cases (particularly, homophobic/transphobic hate speech). We make our code and models publicly available for future research.

Exploring Large Language Models for Hate Speech Detection in Rioplatense Spanish

The proliferation of transliterated texts in digital spaces has emphasized the need for detecting and classifying hate speech in languages beyond English, particularly in low-resource languages. As online discourse can perpetuate discrimination based on target groups, e.g. gender, religion, and origin, multi-label classification of hateful content can help in understanding hate motivation and enhance content moderation. While previous efforts have focused on monolingual or binary hate classification tasks, no work has yet addressed the challenge of multi-label hate speech classification in transliterated Bangla. We introduce BanTH, the first multi-label transliterated Bangla hate speech dataset. The samples are sourced from YouTube comments, where each instance is labeled with one or more target groups, reflecting the regional demographic. We propose a novel translation-based LLM prompting strategy that translates or transliterates under-resourced text to higher-resourced text before classifying the hate group(s). Experiments reveal further pre-trained encoders achieving state-of-the-art performance on the BanTH dataset while translation-based prompting outperforms other strategies in the zero-shot setting. We address a critical gap in Bangla hate speech and set the stage for further exploration into code-mixed and multi-label classification in underrepresented languages.

BanTH: A Multi-label Hate Speech Detection Dataset for Transliterated Bangla

We present THOUGHTSCULPT, a general reasoning and search method for tasks with outputs that can be decomposed into components. THOUGHTSCULPT explores a search tree of potential solutions using Monte Carlo Tree Search (MCTS), building solutions one action at a time and evaluating according to any domain-specific heuristic, which in practice is often simply an LLM evaluator. Critically, our action space includes revision actions: THOUGHTSCULPT may choose to revise part of its previous output rather than continuing to build the rest of its output. Empirically, THOUGHTSCULPT outperforms state-of-the-art reasoning methods across three challenging tasks: Story Outline Improvement (up to +30% interestingness), Mini-Crosswords Solving (up to +16% word success rate), and Constrained Generation (up to +10% concept coverage).

ThoughtSculpt: Reasoning with Intermediate Revision and Search

We introduce methods to quantify how Large Language Models (LLMs) encode and store contextual information, revealing that tokens often seen as minor (e.g., determiners, punctuation) carry surprisingly high context. Notably, removing these tokens — especially stopwords, articles, and commas — consistently degrades performance on MMLU and BABILong-4k, even if removing only irrelevant tokens. Our analysis also shows a strong correlation between contextualization and linearity, where linearity measures how closely the transformation from one layer’s embeddings to the next can be approximated by a single linear mapping. These findings underscore the hidden importance of ``filler'' tokens in maintaining context. For further exploration, we present LLM-Microscope, an open-source toolkit that assesses token-level nonlinearity, evaluates contextual memory, visualizes intermediate layer contributions (via an adapted Logit Lens), and measures the intrinsic dimensionality of representations. This toolkit illuminates how seemingly trivial tokens can be critical for long-range understanding.

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

This paper introduces an LLM-based Latin-to-English translation platform designed to address the challenges of translating Latin texts. We named the model LITERA, which stands for Latin Interpretation and Translations into English for Research Assistance. Through a multi-layered translation process utilizing a fine-tuned version of GPT-4o-mini and GPT-4o, LITERA offers an unprecedented level of accuracy, showcased by greatly improved BLEU scores, particularly in classical Latin, along with improved BLEURT scores. The development of LITERA involved close collaboration with Duke University’s Classical Studies Department, which was instrumental in creating a small, high-quality parallel Latin-English dataset. This paper details the architecture, fine-tuning methodology, and prompting strategies used in LITERA, emphasizing its ability to produce literal translations.

Downloads

Next from NAACL 2025

Text Annotation via Inductive Coding: Comparing Human Experts to LLMs in Qualitative Data Analysis

Similar lecture

Optimising Factual Consistency in Summarisation via Preference Learning from Multiple Imperfect Metrics

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES