Morocco

In-context learning (ICL) in large language models (LLMs) has been shown to operate through task vectors&#39; the representation that summarizes the mapping induced by in-context demonstrations and can be composed by simple arithmetic operations. While this phenomenon is well studied in LLMs, its extension to vision-language models (VLMs) remains underexplored. In this work, we systematically examine the additive compositionality of in-context task vectors in VLMs, extracted from text-side hidden representations. Specifically, we construct compositional visual reasoning tasks with clearly defined subtasks and extract task vectors from few-shot demonstrations. Empirical experiments show that the vector for a complex task can be approximated by adding the vectors of its constituent subtasks. Beyond this, we analyze token-level contextual embeddings and show that additive composition arises because complex-task representations emerge as the superposition of atomic subtask components, preserving semantic structure within the model&#39;s activation space.

EACL 2026

On the Additive Compositionality of Task Vectors in Vision-Language Models

task vector

in-context learning

vision-language model

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. <br>
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) <br><br><br>
<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>
<br><br><br><br>

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

ACL 2026

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

In-context learning (ICL) in large language models (LLMs) has been shown to operate through task vectors' the representation that summarizes the mapping induced by in-context demonstrations and can be composed by simple arithmetic operations. While this phenomenon is well studied in LLMs, its extension to vision-language models (VLMs) remains underexplored. In this work, we systematically examine the additive compositionality of in-context task vectors in VLMs, extracted from text-side hidden representations. Specifically, we construct compositional visual reasoning tasks with clearly defined subtasks and extract task vectors from few-shot demonstrations. Empirical experiments show that the vector for a complex task can be approximated by adding the vectors of its constituent subtasks. Beyond this, we analyze token-level contextual embeddings and show that additive composition arises because complex-task representations emerge as the superposition of atomic subtask components, preserving semantic structure within the model's activation space.

poster

#### *Message from the General Chair, Aline Villavicencio*
I’m delighted and honoured to welcome you to the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026), taking place in the beautiful city of Rabat, in Morocco, in March 24-29, 2026. EACL is the flagship European conference of the Association and EACL 2026 proudly continues our field’s tradition of excellence in scholarship, innovation, and inclusivity. I am deeply grateful to the many volunteers whose dedication, generosity, and tireless efforts have made this conference possible.
For the first time EACL is being hosted in the African continent. This is an important milestone for our community, and we are grateful to our Moroccan hosts for enabling this historic moment by bringing this edition of EACL to Rabat. We are also delighted that the Second Arabic NLP School is co-located with EACL. We hope attendees enjoy this wonderful opportunity to strengthen ties with the Computational Linguistics communities across the African continent. *[Read full message](https://drive.google.com/file/d/14NlmHvwM6fPJuMmOvVh7K0vtQbEyv3SZ/view?usp=sharing)*<br><br>

<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://underline.io/events/522/reception','_blank')">Go to Workshops and Tutorials Program</button></html>
<br><br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to EACL 2026". 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 information**](https://2026.eacl.org/registration/) first.

**Online Registration Form**: https://acl.swoogo.com/eacl2026

Welcome to the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL). EACL 2026 will be held in Rabat, Morocco, from March 24–29, 2026. 

Multimodal Large Language Models (MLLMs) enhance the potential of natural language processing. However, their actual impact on document information extraction remains unclear. In particular, it is unclear whether an MLLM-only pipeline—while simpler—can truly match the performance of traditional OCR+MLLM setups. In this paper, we conduct a large-scale benchmarking study that evaluates various out-of-the-box MLLMs on business-document information extraction. To examine and explore failure modes, we propose an automated hierarchical error analysis framework that leverages large language models (LLMs) to diagnose error patterns systematically. Our findings suggest that OCR may not be necessary for powerful MLLMs, as image-only input can achieve comparable performance to OCR-enhanced approaches. Moreover, we demonstrate that carefully designed schema, exemplars, and instructions can further enhance MLLMs performance. We hope this work can offer practical guidance and valuable insight for advancing document information extraction.

OCR or Not? Rethinking Document Information Extraction in the MLLMs Era with Real-World Large-Scale Datasets

Large Language Models (LLMs) are increasingly used to simulate public opinion and other social phenomena. Most current studies constrain these simulations to multiple-choice or short-answer formats for ease of scoring and comparison, but such closed designs overlook the inherently generative nature of LLMs. In this position paper, we argue that open-endedness, using free-form text that captures topics, viewpoints, and reasoning processes "in" LLMs, is essential for realistic social simulation. Drawing on decades of survey methodology research and recent advances in NLP, we argue why this open-endedness is valuable in LLM social simulations, showing how it can improve measurement and design, support exploration of unanticipated views, and reduce researcher-imposed directive bias. It also captures expressiveness and individuality, aids in pretesting, and ultimately enhances methodological utility. We call for novel practices and evaluation frameworks that leverage rather than constrain the open-ended generative diversity of LLMs, creating synergies between NLP and social science.

Too Open for Opinion? Embracing Open-Endedness in Large Language Models for Social Simulation

Text-to-Visualization (Text2Vis) systems translate natural language queries over tabular data into concise answers and executable visualizations. While closed-source LLMs generate functional code, the resulting charts often lack semantic alignment and clarity—qualities that can only be assessed post-execution. Open-source models struggle even more, frequently producing non-executable or visually poor outputs. Although supervised fine-tuning can improve code executability, it fails to enhance overall visualization quality, as traditional SFT loss cannot capture post-execution feedback. To address this gap, we propose RL-Text2Vis, the first reinforcement learning framework for Text2Vis generation. Built on Group Relative Policy Optimization (GRPO), our method uses a novel multi-objective reward that jointly optimizes textual accuracy, code validity, and visualization quality using post-execution feedback. By training Qwen2.5 models (7B and 14B), RL-Text2Vis achieves a 22% relative improvement in chart quality over GPT-4o on the Text2Vis benchmark and boosts code execution success from 78% to 97% relative to its zero-shot baseline. Our models significantly outperform strong zero-shot and supervised baselines and also demonstrate robust generalization to out-of-domain datasets like VIS-Eval and NVBench. These results establish GRPO as an effective strategy for structured, multimodal reasoning in visualization generation. We release our code at https://github.com/vis-nlp/RL-Text2Vis.

Aligning Text, Code, and Vision: A Multi-Objective Reinforcement Learning Framework for Text-to-Visualization

In science, promotional language ('hype') is increasing and can undermine objective evaluation of evidence, impede research development, and erode trust in science. In this paper, we introduce the task of automatic detection of hype, which we define as hyperbolic or subjective language that authors use to glamorize, promote, embellish, or exaggerate aspects of their research. We propose formalized guidelines for identifying hype language and apply them to annotate a portion of the National Institutes of Health (NIH) grant application corpus. We then evaluate traditional text classifiers and language models on this task, comparing their performance with a human baseline. Our experiments show that formalizing annotation guidelines can help humans reliably annotate candidate hype adjectives and that using our annotated dataset to train machine learning models yields promising results. Our findings highlight the linguistic complexity of the task and the potential need for domain knowledge. While some linguistic works address hype detection, to the best of our knowledge, we are the first to approach it as a natural language processing task.

Hype or not? Formalizing Automatic Promotional Language Detection in Biomedical Research

Advances in large language models (LLMs) have enabled significant capabilities in audio processing, resulting in state-of-the-art models now known as Large Audio Language Models (LALMs). However, minimal work has been done to measure audio understanding beyond automatic speech recognition (ASR). This paper closes that gap by proposing a benchmark suite, SCENEBench (Spatial, Cross-lingual, Environmental, Non-speech Evaluation), that targets a broad form of audio comprehension across four real-world categories: background sound understanding, noise localization, cross-linguistic speech understanding, and vocal characterizer recognition. In addition to performance, we also measure model latency. The purpose of this benchmark suite is to assess the audio beyond just what words are said--- rather, in how they are said and the non-speech components of the audio. To strengthen ecological validity, we include a small human-recorded evaluation split per category. Based on the needs articulated by audio understanding use-cases of accessibility technology and industrial noise monitoring, this benchmark reveals critical gaps in current LALMs. The performance in each task is quite varied, with some tasks having performance far below random chance and others with high accuracy. We also provide a structured error taxonomy to characterize standard failure modes across tasks. These results provide direction for targeted improvements in model capabilities.

SCENEBench: An Audio Understanding Benchmark Grounded in Assistive and Industrial Use Cases

Large-scale Vision Language Models (LVLMs) are increasingly applied to a wide range of real-world multimodal tasks, involving complex visual and linguistic reasoning. As these models become more integrated as practical application, they are expected to handle complex aspects of human interaction. Among these, color perception is a fundamental yet highly variable aspect of visual understanding. It diverges across individuals due to biological factors such as Color Vision Deficiencies (CVDs), as well as differences in culture and language. Despite its importance, perceptual diversity has received limited attention. In our study, we use the Ishihara Test, a widely adopted diagnostic tool for detecting CVDs, to investigate the ability of LVLMs to handle individual-level perceptual diversity from three complementary perspectives: generation, confidence, and internal representation. Our experimental analysis shows that while LVLMs possess linguistic factual knowledge about the Ishihara Test and CVDs and can explain them, they still fail to simulate how people with CVDs perceive colors. These findings highlight the need for multimodal systems that account for color perceptual diversity and broader discussions on inclusiveness and fairness in multimodal AI.

Diagnosing Vision Language Models' Perception by Leveraging Human Methods for Color Vision Deficiencies

Group bias in Large Language Models (LLMs) is a well-documented issue, its impact in high-stakes domains such as personalized nutritional advice remains under explored. This study introduces the USChainMains dataset to systematically evaluate LLMs, prompting them with names associated with specific racial and gender groups and rigorously quantifying the healthfulness of the generated meal recommendations against established dietary standards. The findings demonstrate that LLMs systematically recommend meals with significantly higher levels of adverse nutrients for names associated with Black, Hispanic, or male individuals, thereby reflecting and potentially reinforcing detrimental dietary stereotypes. Furthermore, our analysis of two common mitigation strategies reveals their limitations. While model scaling improves overall recommendation healthfulness, it is insufficient to eliminate the healthfulness gap between demographic groups. Notably, while augmented reasoning was effective in mitigating gender bias, it did not mitigate racial disparities. This work underscores the necessity of developing more nuanced, group-aware debiasing techniques to ensure AI-driven systems advance, rather than hinder, health equity.

"Yuki Gets Sushi, David Gets Steak?'': Uncovering Gender and Racial Biases in LLM-Based Meal Recommendations

Psychometric tests are increasingly used to assess psychological constructs in large language models (LLMs). However, it remains unclear whether these tests -- originally developed for humans -- yield meaningful results when applied to LLMs. In this study, we systematically evaluate the reliability and validity of human psychometric tests for three constructs: sexism, racism, and morality. We find moderate reliability across multiple item and prompt variations. Validity is evaluated through both convergent (i.e., testing theory-based inter-test correlations) and ecological approaches (i.e., testing the alignment between tests scores and behavior in real-world downstream tasks). Crucially, we find that psychometric test scores do not align, and in some cases even negatively correlate with, model behavior in downstream tasks, indicating low ecological validity. Our results highlight that systematic evaluations of psychometric tests are essential before interpreting their scores. They also suggest that psychometric tests designed for humans cannot be applied directly to LLMs without adaptation.

Do Psychometric Tests Work for Large Language Models? Evaluation of Tests on Sexism, Racism, and Morality

With the advancement of automated software engineering, the main interest of research is shifting toward more real-world tasks that reflect the daily activities of software engineers. Despite its importance, software migration, a critical task of adapting code for newer environments, has been largely overlooked. In this work, we propose TimeMachine-bench, a benchmark centered on software migration tasks in real-world Python projects. Our benchmark consists of GitHub repositories whose tests begin to fail in response to dependency updates. The construction process is fully automated, enabling live updates of the benchmark. Furthermore, we curated a human-verified subset to guarantee the solvability of the problems. We evaluated the performance of agent-based baselines built on top of two state-of-the-art LLMs on the verified subset. Our results revealed that, while LLMs demonstrate a certain utility in migration tasks, they still face significant challenges, including spurious solutions in long-tail scenarios and unnecessary edits stemming from suboptimal tool-use strategies. We will release our dataset and implementation to accelerate further progress in this field.

TimeMachine-bench: A Benchmark for Evaluating Model Capabilities in Repository-Level Migration Tasks

Reinforcement learning (RL) has been proven to be an effective and robust method for training neural machine translation systems, especially when paired with powerful reward models that accurately assess translation quality.
However, most research has focused on RL methods that use sentence-level feedback, leading to inefficient learning signals
due to the reward sparsity problem – the model receives a single score for the entire sentence.
To address this, we propose a novel approach that leverages fine-grained, token-level quality assessments along with error severity levels using RL methods. Specifically, we use xCOMET, a state-of-the-art quality estimation system, as our token-level reward model.
We conduct experiments on small and large translation datasets with standard encoder-decoder and large language models-based machine translation systems, comparing the impact of sentence-level versus fine-grained reward signals on translation quality.
Our results show that training with token-level rewards improves translation quality across language pairs over baselines according to both automatic and human evaluation. Furthermore, token-level reward optimization improves training stability, evidenced by a steady increase in mean rewards over training epochs.

Downloads

Next from EACL 2026

OCR or Not? Rethinking Document Information Extraction in the MLLMs Era with Real-World Large-Scale Datasets

Similar lecture

One Task Vector is not Enough: A Large-Scale Study for In-Context Learning