United States

Recent advances in Large Language Models (LLM) have led to substantial interest in their application to commonsense reasoning tasks. Despite their potential, LLMs are susceptible to reasoning errors and hallucinations that may be harmful in use cases where accurate reasoning is critical. This challenge underscores the need for verifiable, debuggable, and repairable LLM reasoning. Recent works have made progress toward verifiable reasoning with LLMs by using them as either (i) a reasoner over an axiomatic knowledge base, or (ii) a semantic parser for use in existing logical inference systems. However, both settings are unable to extract commonsense axioms from the LLM that are not already formalized in the knowledge base, and also lack a reliable method to repair missed commonsense inferences. In this work, we present LLM-TRes, a logical reasoning framework based on the notion of &quot;theory resolution&quot; that allows for seamless integration of the commonsense knowledge from LLMs with a verifiable logical reasoning framework that mitigates hallucinations and facilitates debugging of the reasoning procedure as well as repair. We crucially prove that repaired axioms are theoretically guaranteed to be given precedence over flawed ones in our theory resolution inference process. We conclude by evaluating on three diverse language-based reasoning tasks—preference reasoning, deductive reasoning, and causal commonsense reasoning—and demonstrate the superior performance of LLM-TRes vs. state-of-the-art LLM-based reasoning methods in terms of both accuracy and reasoning correctness.

EMNLP 2024

Verifiable, Debuggable, and Repairable Commonsense Logical Reasoning via LLM-based Theory Resolution

llms

cross-lingual

evaluation

multilingual

### 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!

NAACL 2025

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

Recent advances in Large Language Models (LLM) have led to substantial interest in their application to commonsense reasoning tasks. Despite their potential, LLMs are susceptible to reasoning errors and hallucinations that may be harmful in use cases where accurate reasoning is critical. This challenge underscores the need for verifiable, debuggable, and repairable LLM reasoning. Recent works have made progress toward verifiable reasoning with LLMs by using them as either (i) a reasoner over an axiomatic knowledge base, or (ii) a semantic parser for use in existing logical inference systems. However, both settings are unable to extract commonsense axioms from the LLM that are not already formalized in the knowledge base, and also lack a reliable method to repair missed commonsense inferences. In this work, we present LLM-TRes, a logical reasoning framework based on the notion of "theory resolution" that allows for seamless integration of the commonsense knowledge from LLMs with a verifiable logical reasoning framework that mitigates hallucinations and facilitates debugging of the reasoning procedure as well as repair. We crucially prove that repaired axioms are theoretically guaranteed to be given precedence over flawed ones in our theory resolution inference process. We conclude by evaluating on three diverse language-based reasoning tasks—preference reasoning, deductive reasoning, and causal commonsense reasoning—and demonstrate the superior performance of LLM-TRes vs. state-of-the-art LLM-based reasoning methods in terms of both accuracy and reasoning correctness.

technical paper

## 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 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.

Understanding how Transformer-based Language Models (LMs) learn and recall information is a key goal of the deep learning community. Recent interpretability methods project weights and hidden states obtained from the forward pass to the models' vocabularies, helping to uncover how information flows within LMs. In this work, we extend this methodology to LMs’ backward pass and gradients. We first prove that a gradient matrix can be cast as a low-rank linear combination of its forward and backward passes' inputs. We then develop methods to project these gradients into vocabulary items and explore the mechanics of how new information is stored in the LMs' neurons.

Backward Lens: Projecting Language Model Gradients into the Vocabulary Space

Large language models (LLMs) usually fall short on information extraction (IE) tasks and struggle to follow the complex instructions of IE tasks. This primarily arises from LLMs not being aligned with humans, as mainstream alignment datasets typically do not include IE data. In this paper, we introduce **ADELIE** (**A**ligning large language mo**DEL**s on **I**nformation **E**xtraction), an aligned LLM that effectively solves various IE tasks, including closed IE, open IE, and on-demand IE. We first collect and construct a high-quality alignment corpus IEInstruct for IE. Then we train $ADELIE_{SFT}$ using instruction tuning on IEInstruct. We further train $ADELIE_{SFT}$
with direct preference optimization (DPO) objective, resulting in $ADELIE_{DPO}$. Extensive experiments on various held-out IE datasets demonstrate that our models ($ADELIE_{SFT}$ and $ADELIE_{DPO}$) achieve state-of-the-art (SoTA) performance among open-source models. We further explore the general capabilities of ADELIE, and experimental results reveal that their general capabilities do not exhibit a noticeable decline. We have released the code, data, and models to facilitate further research.

ADELIE: Aligning Large Language Models on Information Extraction

Given the rise of multimedia content, human translators increasingly focus on culturally adapting not only words but also other modalities such as images to convey the same meaning. While several applications stand to benefit from this, machine translation systems remain confined to dealing with language in speech and text. In this work, we introduce a new task of translating images to make them culturally relevant. First, we build three pipelines comprising state-of-the-art generative models to do the task. Next, we build a two-part evaluation dataset -- (i) concept: comprising 600 images that are cross-culturally coherent, focusing on a single concept per image; and (ii) application: comprising 100 images curated from real-world applications. We conduct a multi-faceted human evaluation of translated images to assess for cultural relevance and meaning preservation. We find that as of today, image-editing models fail at this task, but can be improved by leveraging LLMs and retrievers in the loop. Best pipelines can only translate 5% of images for some countries in the easier concept dataset and no translation is successful for some countries in the application dataset, highlighting the challenging nature of the task. Our project webpage is here: https://machine-transcreation.github.io/image-transcreation and our code, data and model outputs can be found here: https://github.com/simran-khanuja/image-transcreation.

An image speaks a thousand words, but can everyone listen? On image transcreation for cultural relevance

Methods for mitigating toxic content through masking and infilling often overlook the decision-making process, leading to either insufficient or excessive modifications of toxic tokens. To address this challenge, we propose XDetox, a novel method that integrates token-level toxicity explanations with the masking and infilling detoxification process. We utilized this approach with two strategies to enhance the performance of detoxification. First, identifying toxic tokens to improve the quality of masking. Second, selecting the regenerated sentence by re-ranking the least toxic sentence among candidates. Our experimental results show state-of-the-art performance across four datasets compared to existing detoxification methods. Furthermore, human evaluations indicate that our method outperforms baselines in both fluency and toxicity reduction. These results demonstrate the effectiveness of our method in text detoxification.

XDetox: Text Detoxification with Token-Level Toxicity Explanations

The rapid growth of evaluation methodologies and datasets for large language models (LLMs) has created a pressing need for their unified integration. Meanwhile, concerns about data contamination and bias compromise the trustworthiness of evaluation findings, while the efficiency of evaluation processes remains a bottleneck due to the significant computational costs associated with LLM inference. In response to these challenges, we introduce FreeEval, a modular framework not only for conducting trustworthy and efficient automatic evaluations of LLMs but also serving as a platform to develop and validate new evaluation methodologies. FreeEval addresses key challenges through: (1) unified abstractions that simplify the integration of diverse evaluation methods, including dynamic evaluations requiring complex LLM interactions; (2) built-in meta-evaluation techniques such as data contamination detection and human evaluation to enhance result fairness; (3) a high-performance infrastructure with distributed computation and caching strategies for efficient large-scale evaluations; and (4) an interactive Visualizer for result analysis and interpretation to support innovation of evaluation techniques. We open-source all our code at https://github.com/WisdomShell/FreeEval and our demostration video, live demo, installation guides are available at: https://freeeval.zhuohao.me/.

FreeEval: A Modular Framework for Trustworthy and Efficient Evaluation of Large Language Models

Recent large language models (LLMs) have enabled the development of advanced agentic systems that can integrate various tools and APIs to fulfill user queries through function calling. However, the deployment of these LLMs on the edge has not been explored since they typically require cloud-based infrastructure due to their substantial model size and computational demands. To this end, we present TinyAgent, an end-to-end framework for training and deploying task-specific small language model agents capable of function calling for driving agentic systems at the edge. We first show how to enable accurate function calling for open-source models via the LLMCompiler framework. We then systematically curate a high-quality dataset for function calling, which we use to fine-tune two small language models, TinyAgent-1.1B and 7B. For efficient inference, we introduce a novel tool retrieval method to reduce the input prompt length and utilize quantization to further accelerate the inference speed. As a driving application, we demonstrate a local Siri-like system for Apple's MacBook that can execute user commands through text or voice input. Our results show that our models can achieve, and even surpass, the function-calling capabilities of larger models like GPT-4-Turbo, while being fully deployed at the edge. We open-source our [dataset, models, and installable package](https://github.com/SqueezeAILab/TinyAgent) and provide a [demo video](https://www.youtube.com/watch?v=0GvaGL9IDpQ) for our MacBook assistant agent.

TinyAgent: Function Calling at the Edge

Recently, there has been increasing interest in using Large Language Models (LLMs) to construct complex multi-agent systems to perform tasks such as compiling literature reviews, drafting consumer reports, and planning vacations. Many tools and libraries exist for helping create such systems, however none support *recursive* multi-agent systems---where the models themselves flexibly decide when to delegate tasks and how to organize their delegation structure. In this work, we introduce ReDel: a toolkit for recursive multi-agent systems that supports custom tool-use, delegation schemes, event-based logging, and interactive replay in an easy-to-use web interface. We show that, using ReDel, we are able to achieve significant performance gains on agentic benchmarks and easily identify potential areas of improvements through the visualization and debugging tools. Our code, documentation, and PyPI package are open-source at https://github.com/zhudotexe/redel, and free to use under the MIT license.

ReDel: A Toolkit for LLM-Powered Recursive Multi-Agent Systems

Large Language Models (LLMs) have significantly advanced QA tasks through in-context learning but often suffer from hallucinations. Attributing supporting evidence grounded in source documents has been explored for unstructured text in the past. However, tabular data present unique challenges for attribution due to ambiguities (e.g., abbreviations, domain-specific terms), complex header hierarchies, and the difficulty in interpreting individual table cells without row and column context. We introduce a new task, Fine-grained Structured Table Attribution (FAST-Tab), to generate row and column-level attributions supporting LLM-generated answers. We present MATSA, a novel LLM-based Multi-Agent system capable of post-hoc Table Structure Attribution to help users visually interpret factual claims derived from tables. MATSA augments tabular entities with descriptive context about structure, metadata, and numerical trends to semantically retrieve relevant rows and columns corresponding to facts in an answer. Additionally, we propose TabCite, a diverse benchmark designed to evaluate the FAST-Tab task on tables with complex layouts sourced from Wikipedia and business PDF documents. Extensive experiments demonstrate that MATSA significantly outperforms SOTA baselines on TabCite, achieving an 8-13% improvement in F1 score. Qualitative user studies show that MATSA helps increase user trust in Generative AI by providing enhanced explainability for LLM-assisted table QA and enables professionals to be more productive by saving time on fact-checking LLM-generated answers.

MATSA: Multi-Agent Table Structure Attribution

We evaluate recent Large Language Models (LLMs) on the challenging task of summarizing short stories, which can be lengthy, and include nuanced subtext or scrambled timelines. Importantly, we work directly with authors to ensure that the stories have not been shared online (and therefore are unseen by the models), and to obtain informed evaluations of summary quality using judgments from the authors themselves. Through quantitative and qualitative analysis grounded in narrative theory, we compare GPT-4, Claude-2.1, and LLama-2-70B. We find that all three models make faithfulness mistakes in over 50% of summaries and struggle with specificity and interpretation of difficult subtext. We additionally demonstrate that LLM ratings and other automatic metrics for summary quality do not correlate well with the quality ratings from the writers.

Reading Subtext: Evaluating Large Language Models on Short Story Summarization with Writers

We deal with the problem of Question Answering (QA) over a long document, which poses a challenge for modern Large Language Models (LLMs). Although LLMs can handle increasingly longer context windows, they struggle to effectively utilize the long content. To address this issue, we introduce the concept of a virtual document (VDoc). A VDoc is created by selecting chunks from the original document that are most likely to contain the information needed to answer the user’s question, while ensuring they fit within the LLM’s context window. We hypothesize that providing a short and focused VDoc to the LLM is more effective than filling the entire context window with less relevant information. Our experiments confirm this hypothesis and demonstrate that using VDocs improves results on the QA task.

Downloads

Next from EMNLP 2024

Backward Lens: Projecting Language Model Gradients into the Vocabulary Space

Similar lecture

Towards Quantifying Commonsense Reasoning with Mechanistic Insights