United States

The escalating challenge of misinformation, particularly in political discourse, requires advanced fact-checking solutions; this is even clearer in the more complex scenario of multimodal claims. We tackle this issue using a multimodal large language model in conjunction with retrieval-augmented generation (RAG), and introduce two novel reasoning techniques: Chain of RAG (CoRAG) and Tree of RAG (ToRAG). They fact-check multimodal claims by extracting both textual and image content, retrieving external information, and reasoning subsequent questions to be answered based on prior evidence. We achieve a weighted F1-score of 0.85, surpassing a baseline reasoning technique by 0.14 points. Human evaluation confirms that the vast majority of our generated fact-check explanations contain all information from gold standard data.

EMNLP 2024

RAGAR, Your Falsehood Radar: RAG-Augmented Reasoning for Political Fact-Checking using Multimodal Large Language Models

reasoning techniques

veracity prediction

gpt-4v

gpt-4

chain-of-thought

llms

factchecking

politics

multimodality

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

ACL 2025

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

## 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.

Please register!

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.

Fact-checking is a critical NLP task aimed at verifying the truthfulness of claims based on evidence and current approaches are labor-intensive and document-centric. FactGenius introduces a novel approach by combining zero-shot prompting of large language models (LLMs) with fuzzy relation mining on Knowledge Graphs (KGs) for fact-checking.

FactGenius: Combining Zero-Shot Prompting and Fuzzy Relation Mining to Improve Fact Verification with Knowledge Graphs

Despite recent success in natural language processing (NLP), fact verification remains a difficult task. Due to misinformation spreading increasingly fast, attention has been directed towards automatically verifying the correctness of claims. In the domain of NLP, this is usually done by training supervised machine learning models to verify claims by utilizing evidence from trustworthy corpora. \textbf{We present efficient methods for verifying claims on a dataset where the evidence is in the form of structured knowledge graphs.} We use the \textsc{FactKG} dataset, which is constructed from the \emph{DBpedia} knowledge graph extracted from Wikipedia. By simplifying the evidence retrieval process, from fine-tuned language models to simple logical retrievals, \textbf{we are able to construct models that both require less computational resources and achieve better test-set accuracy.

Fact or Fiction? Improving Fact Verification with Knowledge Graphs through Simplified Subgraph Retrievals

Contradiction retrieval refers to identifying and extracting documents that explicitly disagree with or refute the content of a query, which is important to many downstream applications like fact checking and data cleaning. To retrieve contradiction argument to the query from large document corpora, existing methods such as similarity search and crossencoder models exhibit significant limitations. The former struggles to capture the essence of contradiction due to its inherent nature of favoring similarity, while the latter suffers from computational inefficiency, especially when the size of corpora is large. To address these challenges, we introduce a novel approach: \name~that leverages specially trained sentence embeddings designed to preserve subtle, contradictory nuances between sentences. Our method utilizes a combined metric of cosine similarity and a sparsity function to efficiently identify and retrieve documents that contradict a given query. This approach dramatically enhances the speed of contradiction detection by reducing the need for exhaustive document comparisons to simple vector calculations. We validate our model using the Arguana dataset, a benchmark dataset specifically geared towards contradiction retrieval, as well as synthetic contradictions generated from the MSMARCO and HotpotQA datasets using GPT-4. Our experiments demonstrate the efficacy of our approach not only in contradiction retrieval with more than 30% accuracy improvements on MSMARCO and HotpotQA across different model architectures but also in applications such as cleaning corrupted corpora to restore high-quality QA retrieval. This paper outlines a promising direction for improving the accuracy and efficiency of contradiction retrieval in large-scale text corpora.

SparseCL: Sparse Contrastive Learning for Contradiction Retrieval

Training automatic summary fact verifiers often faces the challenge of a lack of human-labeled data. Our experimental results reveal that the model trained on extensive LLM-generated datasets surpasses that trained on smaller human-annotated datasets when evaluated using human-generated test sets. Fine-tuning fact verification models with LLM feedback can be more effective and cost-efficient than human feedback.

Learning to Verify Summary Facts with Fine-Grained LLM Feedback

We introduce \textbf{DAHL}, a benchmark dataset and automated evaluation system designed to assess hallucination in long-form text generation, specifically within the biomedical domain. Our benchmark dataset, meticulously curated from biomedical research papers, consists of 8,573 questions across 29 categories. DAHL evaluates fact-conflicting hallucinations in Large Language Models (LLMs) by deconstructing responses into atomic units, each representing a single piece of information. The accuracy of these responses is averaged to produce the \textit{DAHL Score}, offering a more in-depth evaluation of hallucinations compared to previous methods that rely on multiple-choice tasks. We conduct experiments with 8 different models, finding that larger models tend to hallucinate less; however, beyond a model size of 7 to 8 billion parameters, further scaling does not significantly improve factual accuracy. The \textit{DAHL Score} holds potential as an efficient alternative to human-annotated preference labels, being able to be expanded to other specialized domains. We release the dataset and code in [https://github.com/seemdog/DAHL].

DAHL: Domain-specific Automated Hallucination Evaluation of Long-Form Text through a Benchmark Dataset in Biomedicine

Tables play a crucial role in conveying information in various domains. We propose a Plan-then-Reason framework to answer different types of user queries over tables with sentence context. The framework first plans the reasoning paths over the context, then assigns each step to program-based or textual reasoning to reach the final answer. This framework enhances the table reasoning abilities for both in-context learning and fine-tuning methods. GPT-3.5-Turbo following Plan-then-Reason framework surpasses other prompting baselines without self-consistency while using less API calls and in-context demonstrations. We also construct an instruction tuning set TrixInstruct to evaluate the effectiveness of fine-tuning with this framework. We present ProTrix model family by finetuning models on TrixInstruct. Our experiments show that ProTrix family generalizes to diverse unseen tabular tasks with only 6k training instances. We further demonstrate that ProTrix can generate accurate and faithful explanations to answer complex free-form questions. Our work underscores the importance of the planning and reasoning abilities towards a model over tabular tasks with generalizability and interpretability. We will open-source our dataset and models.

ProTrix: Building Models for Planning and Reasoning over Tables with Sentence Context

Misinformation on social media can stem from truthful news articles that are cited in misleading ways. In contrast to traditional media, the collaborative nature of information propagation on these platforms means that facts presented on posts are known second-hand or third-hand, and as such often contain transformations of information that may be misleading or incomplete, potentially leading to a propagation of falsehoods. As a first step in analyzing misinformation occurrences in information transfer scenarios on social media, we propose the task of classifying social media posts that quote news articles from the perspective of information transformation. Specifically, the task is composed of classifying the posts based on the faithfulness of factual claims and the degree to which the posts are misleading. On the basis of the proposed annotation criteria, 200 tweets (X posts) quoting news articles on COVID-19 were labeled by 3 annotators, with the result that 20% of posts were labeled as contained potentially misleading information concerning the article by a majority of annotators. We conduct exploratory experiments leveraging large language models (LMs) to classify information transformation with few-shot prompting, with performance on misleadingness classification approaching that of some human annotators.

Detecting Misleading News Representations on Social Media Posts

Detecting and tracking emerging trends and weak signals in large, evolving text corpora is vital for applications such as monitoring scientific literature, managing brand reputation, surveilling critical infrastructure and more generally to any kind of text-based event detection. Existing solutions often fail to capture the nuanced context or dynamically track evolving patterns over time. BERTrend, a novel method, addresses these limitations using neural topic modeling in an online setting. It introduces a new metric to quantify topic popularity over time by considering both the number of documents and update frequency. This metric classifies topics as noise, weak, or strong signals, flagging emerging, rapidly growing topics for further investigation. Experimentation on two large real-world datasets demonstrates BERTrend's ability to accurately detect and track meaningful weak signals while filtering out noise, offering a comprehensive solution for monitoring emerging trends in large-scale, evolving text corpora. The method can also be used for retrospective analysis of past events. In addition, the use of Large Language Models together with BERTrend offers efficient means for the interpretability of trends of events.

BERTrend: Neural Topic Modeling for Emerging Trends Detection

Event detection in text streams is a crucial task for the analysis of online media and social networks. One of the current challenges in this field is establishing a performance standard while maintaining an acceptable level of computational complexity. In our study, we use an incremental clustering algorithm combined with recent advancements in sentence embeddings. Our objective is to compare our findings with previous studies, specifically those by Cao et al. (2024) and Mazoyer et al. (2020). Our results demonstrate significant improvements and could serve as a relevant baseline for future research in this area.

An Incremental Clustering Baseline for Event Detection on Twitter

Language models (LMs) show exceptional promise in the area of few-shot event extraction, but they suffer from certain limitations. In particular, DEGREE (Hsu et al., 2022) is an LM-based event extraction model that has recently been supplanted by other large language model-based state-of-the-art systems, but it suffers from an inability to cope with multiple events in the same region of an input document. In this work, we present a simple method for extending this system with the ability to gracefully handle different densities of events within documents, thereby rendering it competitive with the state-of-the-art once more, and additionally explore a novel evaluation metric that can be used to qualitatively compare the outputs of different event extraction systems. Finally, we show that our extension allows models to break apart documents into less small pieces during processing without sacrificing accuracy.

Downloads

Next from EMNLP 2024

FactGenius: Combining Zero-Shot Prompting and Fuzzy Relation Mining to Improve Fact Verification with Knowledge Graphs

Similar lecture

MultiReflect: Multimodal Self-Reflective RAG-based Automated Fact-Checking