Austria

We present GaRAGe, a large RAG benchmark with human-curated long-form answers and annotations of each grounding passage, allowing a fine-grained evaluation of whether LLMs can identify relevant grounding when generating RAG answers. Our benchmark contains 2366 questions of diverse complexity, dynamism, and topics, and includes over 35K annotated passages retrieved from both private document sets and the Web, to reflect real-world RAG use cases. This makes it an ideal test bed to evaluate an LLM’s ability to identify only the relevant information necessary to compose a response, or provide a deflective response when there is insufficient information. Evaluations of multiple state-of-the-art LLMs on GaRAGe show that the models tend to over-summarise rather than (a) ground their answers strictly on the annotated relevant passages (reaching at most a Relevance-Aware Factuality Score of 60%), or (b) deflect when no relevant grounding is available (reaching at most 31% true positive rate in deflections). The F1 in attribution to relevant sources is at most 58.9%, and we show that performance is particularly reduced when answering time-sensitive questions and when having to draw knowledge from sparser private grounding sources.

ACL 2025

GaRAGe: A Benchmark with Grounding Annotations for RAG Evaluation

rag benchmark evaluation factuality

poster

### 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 are required to register.
Please complete your registration to continue.

We recommend reading [**the registration details**](https://2025.aclweb.org/registration/) first.

Registration Required

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

Improving context faithfulness in large language models is essential for developing trustworthy retrieval augmented generation systems and mitigating hallucinations, especially in long-form question answering (LFQA) tasks or scenarios involving knowledge conflicts. Existing methods either intervene LLMs only at inference without addressing their inherent limitations or overlook the potential for self-improvement. In this paper, we introduce GenDiE(Generate, Discriminate, Evolve), a novel self-evolving framework that enhances context faithfulness through fine-grained sentence-level optimization. GenDiE combines both generative and discriminative training, equipping LLMs with self-generation and self-scoring capabilities to facilitate iterative self-evolution. This supports both data construction for model alignment and score-guided search during inference. Furthermore, by treating each sentence in a response as an independent optimization unit, GenDiE effectively addresses the limitations of previous approaches that optimize at the holistic answer level, which may miss unfaithful details. Experiments on ASQA (in-domain LFQA) and ConFiQA (out-of-domain counterfactual QA) datasets demonstrate that GenDiE surpasses various baselines in both faithfulness and correctness, and exhibits robust performance for domain adaptation.

Generate, Discriminate, Evolve: Enhancing Context Faithfulness via Fine-Grained Sentence-Level Self-Evolution

Paraphrasing is rooted in semantics, which makes evaluating paraphrase generation systems hard. Current paraphrase generators are typically evaluated using borrowed metrics from adjacent text-to-text tasks, like machine translation or text summarization. These metrics tend to have ties to the surface form of the reference text. This is not ideal for paraphrases as we typically want variation in the lexicon while persisting semantics. To address this problem, and inspired by learned similarity evaluation on plain text, we propose PAM, a Paraphrase AMR-Centric Evaluation Metric. This metric uses AMR graphs extracted from the input text, which consist of semantic structures agnostic to the text surface form, making the resulting evaluation metric more robust to variations in syntax or lexicon. Additionally, we evaluated PAM on different semantic textual similarity datasets and found that it improves the correlations with human semantic scores when compared to other AMR-based metrics.

PAM: Paraphrase AMR-Centric Evaluation Metric

Recent advances in mechanistic interpretability have highlighted the potential of automating interpretability pipelines in analyzing the latent representations within LLMs. While they may enhance our understanding of internal mechanisms, the field lacks standardized evaluation methods for assessing the validity of discovered features. We attempt to bridge this gap by introducing FADE: Feature Alignment to Description Evaluation, a scalable model-agnostic framework for evaluating feature-description alignment. FADE evaluates alignment across four key metrics - Clarity, Responsiveness, Purity, and Faithfulness - and systematically quantifies the causes for the misalignment of feature and their description. We apply FADE to analyze existing open-source feature descriptions, and assess key components of automated interpretability pipelines, aiming to enhance descriptions quality. Our findings highlight fundamental challenges in generating feature descriptions, particularly for SAEs as compared to MLP neurons, providing insights into the limitations and future directions of automated interpretability.

FADE: Why Bad Descriptions Happen to Good Features

Many current studies focus on extracting tests or treatments when constructing clinical pathways, often neglecting the patient's symptoms and diagnosis, leading to incomplete diagnostic and therapeutic logic. Therefore, this paper aims to extract clinical pathways from electronic medical records that encompass complete diagnostic and therapeutic logic, including temporal information, patient symptoms, diagnosis, and tests or treatments. To achieve this objective, we propose a novel clinical pathway representation: the clinical status pathway. We also design a LLM-based pipeline framework for extracting clinical status pathway from electronic medical records, with the core concept being to improve extraction accuracy by modeling the diagnostic and treatment processes. In our experiments, we apply this framework to construct a comprehensive breast cancer-specific clinical status pathway and evaluate its performance on medical question-answering and decision-support tasks, demonstrating significant improvements over traditional clinical pathways. The code is publicly available at https://anonymous.4open.science/r/EMR2CSP-BD11

EMRs2CSP : Mining Clinical Status Pathway from Electronic Medical Records

Analyzing texts spanning long periods of time is critical for researchers in historical linguistics and related disciplines. However, publicly available corpora suitable for such analyses are scarce. The Project Gutenberg (PG) corpus presents a significant yet underutilized opportunity in this context, due to the absence of accurate temporal metadata. We take advantage of language models and information retrieval to explore four sources of information -- Open Web, Wikipedia, Open Library API, and Project Gutenberg (PG) -- to add missing temporal metadata to PG books. Through 20 experiments employing state-of-the-art Large Language Models (LLMs) and Retrieval Augmented Generation (RAG) methods, we estimate the production years of all PG books. We curate an enriched metadata repository for PG books and propose a refined version of the PG corpus, which includes 53774 books with a total of 3.8 billion tokens in 11 languages, produced between 1600 and 2000. This work provides a new resource for computational linguistics and humanities studies focusing on diachronic analyses. All experiments data are published for the public (https://drive.proton.me/urls/DNQE5562RM#L2ldqrQ88Y6B).

Filling the Temporal Void: Recovering Missing Publication Years in the Project Gutenberg Corpus Using LLMs

Large Language Models (LLMs) are increasingly used in tasks requiring interpretive and inferential accuracy. In this paper, we introduce ExpliCa, a new dataset for evaluating LLMs in explicit causal reasoning. ExpliCa uniquely integrates both causal and temporal relations presented in different linguistic orders and explicitly expressed by linguistic connectives. The dataset is enriched with crowdsourced human acceptability ratings. We tested LLMs on ExpliCa through prompting and perplexity-based metrics. We assessed seven commercial and open-source LLMs, revealing that even top models struggle to reach 0.80 accuracy. Interestingly, models tend to confound temporal relations with causal ones, and their performance is also strongly influenced by the linguistic order of the events. Finally, perplexity-based scores and prompting performance are differently affected by model size.

ExpliCa: Evaluating Explicit Causal Reasoning in Large Language Models

Large Language Models (LLMs) have become powerful tools for algorithm design, leveraging extensive knowledge and contextual reasoning. However, existing methods rely on iterative code generation and performance-based filtering, leading to inefficient exploration and high computational costs. Inspired by butterfly metamorphosis, where organisms undergo structured developmental transitions, Phase I allows larvae to accumulate nutrients, laying the foundation for Phase II, where the pupa transforms into a butterfly. Building on this principle, we introduce Metamorphosis on Prompts (MoP) to iteratively refine task descriptions and Metamorphosis on Algorithms (MoA) to generate more effective solutions. Their nested refinement ensures task-algorithm alignment, systematically improving both task descriptions and algorithmic solutions for more efficient algorithm design. To further enhance efficiency, we incorporate predictor-assisted code evaluation, mimicking natural selection by filtering out weak candidates early and reducing computational costs. Experimental results on TSP, MKP, and CVRP demonstrate that NeRM consistently outperforms state-of-the-art approaches in both performance and efficiency.

Nested-Refinement Metamorphosis: Reflective Evolution of Prompts and Code for Efficient Algorithm Design

Multimodal large language models (MLLMs) are prone to non-factual or outdated knowledge issues, highlighting the importance of knowledge editing. Many benchmark has been proposed for researching multimodal knowledge editing. However, previous benchmarks focus on limited scenarios due to the lack of rigorous definition of multimodal knowledge. To better evaluate multimodal knowledge editing, we propose a decomposed definition of multimodal knowledge. Following the decomposed definition of multimodal knowledge, we introduce three scenarios and a novel requirement modality consistency. We construct MC-MKE, a fine-grained **M**ultimodal **K**nowledge **E**diting benchmark emphasizing **M**odality **C**onsistency through strict data selection. We evaluate four multimodal knowledge editing methods on MC-MKE, revealing their limitations, particularly in terms of modality consistency. Our work highlights the challenges posed by multimodal knowledge editing and motivates further research in developing effective techniques for this task.

MC-MKE: A Fine-Grained Multimodal Knowledge Editing Benchmark Emphasizing Modality Consistency

Fine-tuning LLMs with first-order methods like back-propagation is computationally intensive. Zeroth-Order (ZO) optimisation, using function evaluations instead of gradients, reduces memory usage but suffers from slow convergence in high-dimensional models. As a result, ZO research in LLMs has mostly focused on classification, overlooking more complex generative tasks. In this paper, we introduce ZOPrO, a novel ZO algorithm designed for *Preference Optimisation* in LLMs. We begin by analysing the interplay between policy and reward models during traditional (first-order) Preference Optimisation, uncovering patterns in their relative updates. Guided by these insights, we adapt Simultaneous Perturbation Stochastic Approximation (SPSA) with a targeted sampling strategy to accelerate convergence. Through experiments on summarisation, machine translation, and conversational assistants, we demonstrate that our method consistently enhances reward signals while achieving convergence times comparable to first-order methods. While it falls short of some state-of-the-art methods, our work is the first to apply Zeroth-Order methods to Preference Optimisation in LLMs, going beyond classification tasks and paving the way for a largely unexplored research direction.

Visualising Policy-Reward Interplay to Inform Zeroth-Order Preference Optimisation of Large Language Models

This paper presents a comprehensive evaluation of the capabilities of Large Language Models (LLMs) in metaphor interpretation across multiple datasets, tasks, and prompt configurations. Although metaphor processing has gained significant attention in Natural Language Processing (NLP), previous research has been limited to single-dataset evaluations and specific task settings, often using artificially constructed data through lexical replacement. We address these limitations by conducting extensive experiments using diverse publicly available datasets with inference and metaphor annotations, focusing on Natural Language Inference (NLI) and Question Answering (QA) tasks. The results indicate that LLMs' performance is more influenced by features like lexical overlap and sentence length than by metaphorical content, demonstrating that any alleged emergent abilities of LLMs to understand metaphorical language are the result of a combination of surface-level features, in-context learning, and linguistic knowledge. This work provides critical insights into the current capabilities and limitations of LLMs in processing figurative language, highlighting the need for more realistic evaluation frameworks in metaphor interpretation tasks. Data and code publicly available: https://anonymous.4open.science/r/metaphorLLM-047E/

GaRAGe: A Benchmark with Grounding Annotations for RAG Evaluation

Downloads

Next from ACL 2025

Generate, Discriminate, Evolve: Enhancing Context Faithfulness via Fine-Grained Sentence-Level Self-Evolution

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES