United States

Self-consistency (SC) improves the performance of large language models (LLMs) across various tasks and domains that involve short content. However, does this support its effectiveness for long-context problems?
We challenge the assumption that SC&#39;s benefits generalize to long-context settings, where LLMs often struggle with position bias---the systematic over-reliance on specific context regions---which hinders their ability to utilize information effectively from all parts of their context. Through comprehensive experimentation with varying state-of-the-art models, tasks, and SC formulations, we find that SC not only fails to improve but actively degrades performance on long-context tasks. This degradation is driven by persistent position bias, which worsens with longer context lengths and smaller model sizes but remains invariant to prompt format or task type. Unlike short-context tasks, where SC diversifies reasoning paths, long-context SC amplifies positional errors. These comprehensive results provide valuable insight into the limitations of current LLMs in long-context understanding and highlight the need for more sophisticated approaches.

ACL 2026

Self-Consistency Falls Short!: The Adverse Effects of Positional Bias on Long-Context Problems

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

Self-consistency (SC) improves the performance of large language models (LLMs) across various tasks and domains that involve short content. However, does this support its effectiveness for long-context problems?
We challenge the assumption that SC's benefits generalize to long-context settings, where LLMs often struggle with position bias---the systematic over-reliance on specific context regions---which hinders their ability to utilize information effectively from all parts of their context. Through comprehensive experimentation with varying state-of-the-art models, tasks, and SC formulations, we find that SC not only fails to improve but actively degrades performance on long-context tasks. This degradation is driven by persistent position bias, which worsens with longer context lengths and smaller model sizes but remains invariant to prompt format or task type. Unlike short-context tasks, where SC diversifies reasoning paths, long-context SC amplifies positional errors. These comprehensive results provide valuable insight into the limitations of current LLMs in long-context understanding and highlight the need for more sophisticated approaches.

technical paper

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields 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.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

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 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

**Online Registration Form**: https://acl.swoogo.com/acl2026

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

We propose a novel data synthesis framework, DecIF, which automatically generates accurate and diverse instruction-following data from scratch for supervised fine-tuning (SFT) and reinforcement learning (RL), leveraging large language models (LLMs) and minimal external resources. By decomposing the data synthesis pipeline into fine-grained steps, DecIF achieves meticulous quality and diversity control over generated instruction-following data. Extensive experiments across both SFT and RL demonstrate DecIF’s strong capability to flexibly synthesize accurate instruction-following data for both paradigms compared to comprehensive baselines. Further analysis demonstrates the framework's robustness, scalability, and computational efficiency in instruction-following data generation, while its modular design ensures straightforward implementation and reproducibility.

DecIF: Improving Instruction-Following through Decomposition

Current evaluation paradigms for emotional support conversations tend to reward generic empathetic responses, yet they fail to assess whether the support is genuinely personalized to users’ unique psychological profiles and contextual needs. We introduce \textbf{EmoHarbor}, an automated evaluation framework that adopts a \textbf{User-as-a-Judge} paradigm by simulating the user's inner world. EmoHarbor employs a Chain-of-Agent architecture that decomposes users' internal processes into three specialized roles, enabling agents to interact with supporters and complete assessments in a manner similar to human users. We instantiate this benchmark using 100 real-world user profiles that cover diverse personality traits and situations, and define 10 evaluation dimensions of personalized support quality. Comprehensive evaluation of 20 advanced LLMs on EmoHarbor reveals a critical insight: while these models excel at generating empathetic responses, they consistently fail to tailor support to individual user contexts. This finding reframes the central challenge, shifting research focus from merely enhancing generic empathy to developing truly user-aware emotional support. EmoHarbor provides a reproducible and scalable framework to guide the development and evaluation of more nuanced and user-aware emotional support systems.

EmoHarbor: Evaluating Personalized Emotional Support by Simulating the User's Internal World

Large Language Models (LLMs) often fail to recognize fallacious reasoning in real-world interactions, despite strong performance on static fallacy detection tasks. We define this ability as \textbf{fallacy awareness}, the capacity to autonomously perceive and resist fallacies in dynamic, pragmatic contexts. To study this, we introduce \textbf{ISFallacy}, a large-scale Chinese benchmark of 50K interactive scenarios spanning six fallacy types, five social interaction settings, diverse role relationships, and personality traits. We further propose \textbf{FATE}, a two-stage evaluation framework that assesses fallacy awareness without explicit cues, combining natural dialogue responses and reasoning-based decisions. Experiments on five representative LLMs reveal a substantial gap between fallacy classification and awareness, with models particularly vulnerable to emotion-driven fallacies and scenarios involving cooperative or trust-based relationships. Deeper analysis uncovers a cognition–behavior gap and fragile internal representations underlying awareness failures. Our work establishes a foundation for evaluating and enhancing the robustness of LLMs against fallacious reasoning in interactive settings.

Beyond Detection: Evaluating Fallacy Awareness of LLMs in Interactive Scenarios

Translating brain signals into text could restore communication for people with severe paralysis, yet practically usable systems to date rely on invasive electrocorticography (ECoG). Electroencephalography (EEG) offers a non-invasive alternative, and EEG-to-text (EEG2Text) has been widely explored. Interestingly, however, EEG2Text models generally rely on teacher-forcing evaluation; without it, they fail to generate meaningful decoding. This reliance prevents EEG2Text from being applied in real-world, non-academic settings. This has fueled numerous debates about whether EEG2Text is a meaningful direction, by extension, and whether EEG truly contains decodable linguistic information. Here, using a neuropsychology-informed paradigm, we find that existing EEG2Text benchmarks have neglected EEG instability, a flaw that has confounded inference and sparked debate. Our experiments furnish key evidence for the feasibility of teacher-forcing-free EEG2Text decoding. Accordingly, we assemble the Corpus OF Eeg-To-Text (COFETT) using a 128-channel high-density EEG cap, providing a benchmark dedicated to evaluating EEG2Text models. In comparisons with multiple existing benchmarks, COFETT achieves SOTA ability to distinguish among model performances and enables robust, teacher-forcing-free evaluation, thereby opening a path toward practical EEG2Text applications. COFETT is open sourced in https://github.com/baoyudu/COFETT.

Is EEG-to-Text Feasible in Real-World Scenarios? An In-Depth Analysis Using a Neuropsychology-Inspired Benchmark

Spreadsheets are central to real-world applications such as enterprise reporting, auditing, and scientific data management. Despite their ubiquity, existing large language model based approaches typically treat tables as plain text, overlooking critical layout cues and visual semantics. Moreover, real-world spreadsheets are often massive in scale, exceeding the input length that LLMs can efficiently process. To address these challenges, we propose SpreadsheetAgent, a two-stage multi-agent framework for spreadsheet understanding that adopts a step-by-step reading and reasoning paradigm. Instead of loading the entire spreadsheet at once, SpreadsheetAgent incrementally interprets localized regions through multiple modalities, including code execution results, images, and LaTeX tables. The method first constructs a structural sketch and row/column summaries, and then performs task-driven reasoning over this intermediate representation in the Solving Stage. To further enhance reliability, we design a verification module that validates extracted structures via targeted inspections, reducing error propagation and ensuring trustworthy inputs for downstream reasoning. Extensive experiments on two spreadsheet datasets demonstrate the effectiveness of our approach. With GPT-OSS-120B, SpreadsheetAgent achieves 38.16% on Spreadsheet Bench, outperforming the ChatGPT Agent baseline (35.27%) by 2.89 absolute points. These results highlight the potential of SpreadsheetAgent to advance robust and scalable spreadsheet understanding in real-world applications. Code is available at https://github.com/renhouxing/SpreadsheetAgent.

Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning

In this paper, we investigate knowledge forgetting in large language models with a focus on its generalisation—ensuring that models forget not only specific training samples but also related implicit knowledge. To this end, we begin by identifying a broader unlearning scope that includes both target data and logically associated samples, including rephrased, subject-replaced, relation-reversed, and one-hop reasoned data. We then conduct a rigorous evaluation of 15 state-of-the-art methods across three datasets, revealing that unlearned models still recall paraphrased answers and retain target facts in their intermediate layers. This motivates us to take a preliminary step toward more generalised implicit knowledge forgetting by proposing PERMU—a novel probability perturbation-based unlearning paradigm. PERMU simulates adversarial unlearning samples to eliminate fact-related tokens from the logit distribution, collectively reducing the probabilities of all answer-associated tokens. Experiments are conducted on a diverse range of datasets, including TOFU, Harry Potter, ZsRE, WMDP, and MUSE, using models ranging from 1.3B to 13B in scale. The results demonstrate that PERMU delivers up to a 50.40% improvement in unlearning vanilla target data while maintaining a 40.73% boost in forgetting implicit knowledge. Our code can be found in the supplementary material.

Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models

The development of large-scale Visual Question Answering (VQA) datasets has traditionally relied on resource-intensive manual annotation. In addition, most of the existing Arabic VQA datasets focus on culturally-specific and dialect-aware domains. To address these limitations, we propose a new pipeline that leverages Wikipedia template tags to extract the relevant information for each image, which is subsequently utilized by the Large Language Model (LLM) to synthetically generate a new visual question answering dataset. Using this pipeline, we have constructed AraVQA, the most comprehensive Arabic Factoid Visual Question Answering dataset, containing more than 50,000 questions and covering over 20 varied primary subjects within Arabic general knowledge. Our detailed analysis shows that our dataset can serve as a post-training dataset to enhance the performance of existing Visual Language Models (VLMs) on Arabic VQA tasks. Furthermore, we present a novel benchmark, derived from our dataset and validated through manual annotation, that poses more challenges to Arabic VLMs than existing Arabic VQA datasets.

AraVQA: Building a New Arabic Factoid Visual Question Answering Dataset from Wikipedia

The powerful generative capacity of Large Language Models (LLMs) has instigated a paradigm shift in recommendation. However, existing generative models (e.g., OneRec) operate as implicit predictors, critically lacking the capacity for explicit and controllable reasoning—a key advantage of LLMs. To bridge this gap, we propose OneRec-Think, a unified framework that seamlessly integrates dialogue, reasoning, and personalized recommendation. OneRec-Think incorporates: (1) Itemic Alignment: cross-modal Item-Textual Alignment for semantic grounding; (2) Reasoning Activation: Reasoning Scaffolding to activate LLM reasoning within the recommendation context; and (3) Reasoning Enhancement, where we design a recommendation-specific reward function that accounts for the multi-validity nature of user preferences. Experiments across public benchmarks show state-of-the-art performance. Moreover, our proposed "Think-Ahead" architecture enables effective industrial deployment, achieving a 0.159% gain in APP Stay Time and validating the practical efficacy of the model's explicit reasoning capability.

OneRec-Think: In-Text Reasoning for Generative Recommendation

Retrieval-augmented generation (RAG) over long documents typically involves splitting the text into smaller chunks, which serve as the basic units for retrieval. However, due to dependencies across the original document, contextual information is often essential for accurately interpreting each chunk. To address this, prior work has explored encoding longer context windows to produce embeddings for longer chunks, yet their gains in retrieval and downstream tasks remain limited. This is because (1) longer chunks strain the capacity of embedding models due to the increased amount of information they must encode, and (2) many real-world applications still require returning localized evidence due to constraints on model or human bandwidth. To this end, we propose an alternative approach to this challenge by representing short chunks in a way that is conditioned on a broader context window to enhance retrieval performance – i.e., situating a chunk’s meaning within its context. We further show that existing embedding models are not well-equipped to encode such situated context effectively, and thus introduce a new training paradigm and develop the first situated embedding model. To evaluate our method, we curate a book-plot retrieval dataset specifically designed to assess situated retrieval capabilities. On this benchmark, our 1B-parameter model substantially outperforms state-of-the-art embedding models, including several with up to 7B parameters.

Situated Embedding Models for Context-Aware Dense Retrieval

Long-context modeling is critical for a wide range of real-world tasks, including long-context question answering, summarization, and complex reasoning tasks. Recent studies have explored fine-tuning Large Language Models (LLMs) with synthetic data to enhance their long-context capabilities. However, the effectiveness of such approaches is often limited by the low diversity and factual inconsistencies in the generated data. To address these challenges, we propose LongMab, a novel framework that leverages a Multi-Armed Bandit (MAB) rollout strategy to identify the most informative chunks from the given long context for sampling high-quality and diverse responses and constructing preference data pairs for Direct Preference Optimization (DPO) training. Specifically, we treat context chunks as arms of MAB, select chunks based on their expected reward scores to input into LLMs to generate responses, and iteratively update these scores based on reward feedback. Both exploration and exploitation during the rollout process enable the LLM to focus on the most relevant context segments, thereby generating and collecting high-quality and diverse responses. Experimental results on both Llama and Qwen show the effectiveness of LongMab by achieving more than a 4% improvement on long-context reasoning benchmarks. All data and code will be released on https://github.com/NEUIR/LongMab-PO.

Downloads

Next from ACL 2026

DecIF: Improving Instruction-Following through Decomposition

Similar lecture

Soft Self-Consistency Improves Language Models Agents