Austria

Despite near-perfect results reported in the literature, the effectiveness of model editing in real-world applications remains unclear. To bridge this gap, we introduce QAEdit, a new benchmark aligned with widely used question answering (QA) datasets, and WILD, a task-agnostic evaluation framework designed to better reflect real-world usage of model editing. Our single editing experiments show that current editing methods perform substantially worse than previously reported (38.5% vs. 96.8%). We demonstrate that it stems from issues in the synthetic evaluation practices of prior work. Among them, the most severe is the use of teacher forcing during testing, which leaks both content and length of the ground truth, leading to overestimated performance. Furthermore, we simulate practical deployment by sequential editing, revealing that current approaches fail drastically with only 1000 edits. This work calls for a shift in model editing research toward rigorous evaluation and the development of robust, scalable methods that can reliably update knowledge in LLMs for real-world use.

ACL 2025

The Mirage of Model Editing: Revisiting Evaluation in the Wild

teacher forcing

knowledge update

model editing

### 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**. 

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

poster

Despite their success, large language models (LLMs) suffer from notorious hallucination issue. By introducing external knowledge stored in knowledge graphs (KGs), existing methods use paths as the medium to represent the graph information that send into LLMs. However, paths only contain limited graph structure information and are unorganized with redundant sequentially appeared keywords, which are difficult for LLMs to digest. We aim to find a suitable medium that captures the essence of structure knowledge in KGs. Inspired by the Neural Message Passing in Graph Neural Networks, we propose Language Message Passing (LMP) that first learns a concise facts graph by iteratively aggregates neighbor entities and transforms them into semantic facts, and then we performs Topological Readout that encodes the graph structure information into multi-level lists of texts to augment LLMs. Our method serves as a brand-new innovative framework that brings a new perspective into KG-enhanced LLMs, and also offers human-level semantic explainability with significant performance improvements over existing methods on all 5 knowledge graph question answering datasets. Code is available at https://github.com/wanjunhong0/LMP.

Digest the Knowledge: Large Language Models empowered Message Passing for Knowledge Graph Question Answering

Modern recommender systems aim to deeply understand users' complex preferences through their past interactions. While deep collaborative filtering approaches using Graph Neural Networks (GNNs) excel at capturing user-item relationships, their effectiveness is limited when handling sparse data or zero-shot scenarios, primarily due to constraints in ID-based embedding functions. To address these challenges, we propose a model-agnostic recommendation instruction-tuning paradigm that seamlessly integrates large language models with collaborative filtering. Our proposed Recommendation Language Model (RecLM) enhances the capture of user preference diversity through a carefully designed reinforcement learning reward function that facilitates self-augmentation of language models. Comprehensive evaluations demonstrate significant advantages of our approach across various settings, and its plug-and-play compatibility with state-of-the-art recommender systems results in notable performance enhancements.

RecLM: Recommendation Instruction Tuning

While Large Language Models (LLMs) excel at many natural language processing tasks, they often suffer from factual inaccuracies in knowledge-intensive scenarios. Integrating external knowledge resources, particularly knowledge graphs (KGs), provides a transparent and updatable foundation for more reliable reasoning. Knowledge Base Question Answering (KBQA), which queries and reasons over KGs, is central to this effort, especially for complex, multi-hop queries. However, multi-hop reasoning poses two key challenges: (1) maintaining coherent reasoning paths, and (2) avoiding prematurely discarding critical multi-hop connections. To address these issues, we introduce iQUEST, a question-guided KBQA framework that iteratively decomposes complex queries into simpler sub-questions, ensuring a structured and focused reasoning trajectory. Additionally, we integrate a Graph Neural Network (GNN) to look ahead and incorporate 2-hop neighbor information at each reasoning step. This dual approach strengthens the reasoning process, enabling the model to explore viable paths more effectively. Detailed experiments demonstrate the consistent improvement delivered by iQUEST across four benchmark datasets and four LLMs.

iQUEST: An Iterative Question-Guided Framework for Knowledge Base Question Answering

Topic models and document-clustering evaluations either use automated metrics that align poorly with human preferences, or require expert labels that are intractable to scale. We design a scalable human evaluation protocol and a corresponding automated approximation that reflect practitioners' real-world usage of models. Annotators---or an LLM-based proxy---review text items assigned to a topic or cluster, infer a category for the group, then apply that category to other documents. Using this protocol, we collect extensive crowdworker annotations of outputs from a diverse set of topic models on two datasets. We then use these annotations to validate automated proxies, finding that the best LLM proxy is statistically indistinguishable from a human annotator and can therefore serve as a reasonable substitute in automated evaluations.

ProxAnn: Use-Oriented Evaluations of Topic Models and Document Clustering

Large language models (LLMs) encode vast world knowledge but struggle to stay up-to-date, often leading to errors and hallucinations. Knowledge editing offers an efficient alternative to retraining, enabling targeted modifications by updating specific model parameters. However, existing methods primarily focus on individual models, posing challenges in efficiently updating multiple models and adapting to new models. To address this, we propose OnceEdit, a novel ensemble-based approach that employs a plug-in model as the editing module, enabling stable knowledge updates across multiple models. Building on the model ensemble, OnceEdit introduces two key mechanisms to enhance its effectiveness. First, we introduce a dynamic weight mechanism through a weight token for distinguishing between edit-related and non-edit-related instances, ensuring the appropriate utilization of knowledge from integrated models. Second, we incorporate an ensemble enhancement mechanism to mitigate the excessive reliance on the central model inherent in the model ensemble technique, making it more suitable for knowledge editing. Extensive experiments on diverse LLMs demonstrate that OnceEdit consistently outperforms existing methods while achieving superior editing efficiency. Further analysis confirms its adaptability and stability in multi-model editing scenarios.

One for All: Update Parameterized Knowledge Across Multiple Models with Once Edit

This paper proposes CrisisTS, the first multimodal and multilingual dataset for urgency classification composed of benchmark crisis datasets from French and English social media about various expected (e.g., flood, storm) and sudden (e.g., earthquakes, explosions) crises that have been mapped with open source geocoded meteorological time series data. This mapping is based on a simple and effective strategy that allows for temporal and location alignment even in the absence of location mention in the text. A set of multimodal experiments have been conducted relying on transformers and LLMs to improve overall performances while ensuring model generalizability. Our results show that modality fusion outperforms text-only models.

CrisisTS: Coupling Social Media Textual Data and Meteorological Time Series for Urgency Classification

Large language models (LLMs) are trained on extensive historical corpora, but their ability to understand time and maintain temporal awareness of time-evolving factual knowledge remains limited. Previous studies often neglect the critical aspect of utilizing knowledge from various sources. To address this gap, we introduce EvolveBench, a comprehensive benchmark that evaluates temporal competence along five key dimensions: Cognition, which examines the ability to recall and contextualize historical facts. Awareness, which tests LLMs' awareness of temporal misalignment between external inputs and the temporal context of a query. Trustworthiness, which assesses whether models can identify and appropriately refuse queries based on invalid timestamps. Understanding, which focuses on interpreting both explicit dates and implicit historical markers. Finally, reasoning evaluates the capacity to analyze temporal relationships and draw accurate inferences. Evaluating 15 widely used LLMs on EvolveBench shows that GPT-4o achieves the highest average EM score of 79.36, while the open-source Llama3.1-70B demonstrates notable strength in handling temporally misaligned contexts with an average score of 72.47. Despite these advances, all models still struggle with handling temporal misaligned context. Our code and dataset are available at https://github.com/zzysjtuiwct/EvolveBench.

EvolveBench: A Comprehensive Benchmark for Assessing Temporal Awareness in LLMs on Evolving Knowledge

In this work, we conduct systematic research in a challenging area: the automatic evaluation of book-length stories (>100K tokens). Our study focuses on two key questions: (1) understanding which evaluation aspects matter most to readers, and (2) exploring effective methods for evaluating lengthy stories. We introduce the first large-scale benchmark, **LongStoryEval**, comprising 600 newly published books with an average length of 121K tokens (maximum 397K). Each book includes its average rating and multiple reader reviews, presented as critiques organized by evaluation aspects. By analyzing all user-mentioned aspects, we propose an *evaluation criteria structure* and conduct experiments to identify the most significant aspects among the 8 top-level criteria. For evaluation methods, we compare the effectiveness of three types: *aggregation-based*, *incremental-updated*, and *summary-based* evaluations. Our findings reveal that aggregation- and summary-based evaluations perform better, with the former excelling in detail assessment and the latter offering greater efficiency. Building on these insights, we further propose **NovelCritique**, an 8B model that leverages the efficient summary-based framework to review and score stories across specified aspects. NovelCritique outperforms commercial models like GPT-4o in aligning with human evaluations. All our datasets and codes will be released to foster further research.

What Matters in Evaluating Book-Length Stories? A Systematic Study of Long Story Evaluation

In recent years, multimodal large language models (MLLMs) have made significant progress but continue to face inherent challenges in multimodal reasoning, which requires multi-level (e.g., perception, reasoning) and multi-granular (e.g., multi-step reasoning chain) advanced inferencing. Prior work on estimating model confidence tends to focus on the overall response for training and calibration, but fails to assess confidence in each reasoning step, leading to undesirable hallucination snowballing. In this work, we present MMBoundary, a novel framework that advances the knowledge boundary awareness of MLLMs through reasoning step confidence calibration. To achieve this, we propose to incorporate complementary textual and cross-modal self-rewarding signals to estimate confidence at each step of the MLLM reasoning process. In addition to supervised fine-tuning MLLM on this set of self-rewarding confidence estimation signal for initial confidence expression warm-up, we introduce a reinforcement learning stage with multiple reward functions for further aligning model knowledge and calibrating confidence at each reasoning step, enhancing reasoning chain self-correction. Empirical results show that MMBoundary significantly outperforms existing methods across diverse domain datasets and metrics, achieving an average of 7.5% reduction in multimodal confidence calibration errors and up to 8.3% improvement in task performance.

MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibration

As Large Language Models (LLMs) evolve in natural language processing (NLP), their ability to stably follow instructions in long-context inputs has become critical for real-world applications. However, existing benchmarks seldom focus on instruction-following in long-context scenarios or stability on different inputs. To bridge this gap, we introduce LIFBench, a scalable dataset designed to evaluate LLMs' instruction-following capabilities and stability across long contexts. LIFBench comprises three long-context scenarios and eleven diverse tasks, featuring 2,766 instructions generated through an automated expansion method across three dimensions: length, expression, and variables. For evaluation, we propose LIFEval, a rubric-based assessment method that enables precise, automated scoring of complex LLM responses without reliance on LLM-assisted assessments or human judgment. This method allows for a comprehensive analysis of model performance and stability from multiple perspectives. We conduct detailed experiments on 20 prominent LLMs across six length intervals. Our work contributes LIFBench and LIFEval as robust tools for assessing LLM performance in complex and long-context settings, offering valuable insights to guide future advancements in LLM development.

Downloads

Next from ACL 2025

Digest the Knowledge: Large Language Models empowered Message Passing for Knowledge Graph Question Answering

Similar lecture

The Mirage of Model Editing: Revisiting Evaluation in the Wild