Morocco

Large Language Models (LLMs) struggle with syllogistic reasoning, frequently conflating content plausibility with logical validity. This well-known content effect undermines their capacity to act as reliable deductive reasoners, particularly in multilingual contexts where both linguistic variability and world knowledge may exacerbate biases. Prior work shows that prompting and tuning interventions can alleviate these issues only partially, leaving models vulnerable to semantic interference. To make reasoning more consistent, robust, and transferable across languages, we investigate the use of activation steeringвЂ”an inference-time intervention that modulates internal representations to suppress content effects. Our experiments demonstrate that steering techniques constructed for English-based syllogisms generalise effectively to multilingual datasets, yielding higher formal reasoning accuracy while minimally affecting language modelling performance. Moreover, steering supports partial transfer to out-of-distribution tasks, highlighting its potential as a scalable mechanism for content-invariant reasoning. These findings advance the prospect of developing LLMs that can serve as reliable soft reasoners across language landscapes.

EACL 2026

Can Activation Steering Generalize Across Languages? A Study on Syllogistic Reasoning in Language Models

activation steering

llms

reasoning

## 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

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.

technical paper

#### *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)* 

<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>

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 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

Registration Required

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. 

While mechanistic interpretability has developed powerful tools to analyze the internal workings of Large Language Models (LLMs), their complexity has created an accessibility gap, limiting their use to specialists. We address this challenge by designing, building, and evaluating ELIA (Explainable Language Interpretability Analysis), an interactive web application that simplifies the outcomes of various language model component analyses for a broader audience. The system integrates three key techniques -- Attribution Analysis, Function Vector Analysis, and Circuit Tracing -- and introduces a novel methodology: using a vision-language model to automatically generate natural language explanations (NLEs) for the complex visualizations produced by these methods. The effectiveness of this approach was empirically validated through a mixed-methods user study, which revealed a clear preference for interactive, explorable interfaces over simpler, static visualizations. A key finding was that the AI-powered explanations successfully bridged the knowledge gap for non-experts; a statistical analysis showed no significant correlation between a user's prior LLM experience and their comprehension scores, indicating that the system effectively leveled the playing field. We conclude that an AI system can indeed simplify complex model analyses, but its true power is unlocked when paired with thoughtful, user-centered design that prioritizes interactivity, specificity, and narrative guidance.

Simplifying Outcomes of Language Model Component Analyses with ELIA

Tokenization underlies every large language model, yet it remains an under-theorized and inconsistently designed component. Common subword approaches such as Byte Pair Encoding (BPE) offer scalability but often misalign with linguistic structure, amplify bias, and waste capacity across languages and domains. This paper reframes tokenization as a core modeling decision rather than a preprocessing step. We argue for a context-aware framework that integrates tokenizer and model co-design, guided by linguistic, domain, and deployment considerations. Standardized evaluation and transparent reporting are essential to make tokenization choices accountable and comparable. Treating tokenization as a core design problem, not a technical afterthought, can yield language technologies that are fairer, more efficient, and more adaptable.

Stop Taking Tokenizers for Granted: They Are Core Design Decisions in Large Language Models

Evaluating Large Language Models (LLMs) for mental health support poses unique challenges to reliable evaluation due to the emotionally and cognitively complex nature of therapeutic dialogue. Existing benchmarks are limited in scale, authenticity, and reliability, often relying on synthetic or social media data, and lack frameworks to assess when automated judges can be trusted. To address the need for large-scale authentic dialogue datasets and judge-reliability assessment, we introduce two benchmarks that provide a framework for generation and evaluation in this domain. MentalBench-100k consolidates 10,000 authentic single-session therapeutic conversations from three real-world scenarios datasets, each paired with nine LLM-generated responses, yielding 100,000 response pairs. MentalAlign-70k reframes evaluation by comparing four high-performing LLM judges with human experts across 70,000 ratings on seven attributes, grouped into Cognitive Support Score (CSS) and Affective Resonance Score (ARS). We then employ the AffectiveвЂ“Cognitive Agreement Framework, a statistical methodology using intraclass correlation coefficients (ICC) with confidence intervals to quantify agreement, consistency, and bias between LLM judges and human experts. Our analysis reveals systematic inflation by LLM judges, strong reliability for cognitive attributes such as guidance and informativeness, reduced precision for empathy, and some unreliability in safety and relevance. Our contributions establish new methodological and empirical foundations for the reliable and large-scale evaluation of LLMs in mental health contexts.

When Can We Trust LLMs in Mental Health? Large-Scale Benchmarks for Reliable LLM Evaluation

Retrieving the biological impacts of protein-protein interactions (PPIs) is essential for target identification (Target ID) in drug development. Given the vast number of proteins involved, this process remains time-consuming and challenging. Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) frameworks have supported Target ID; however, no benchmark currently exists for identifying the biological impacts of PPIs. To bridge this gap, we introduce the RAG Benchmark for PPIs (RAGPPI), a factual question-answer benchmark of 4,420 question-answer pairs that focus on the potential biological impacts of PPIs. Through interviews with experts, we identified criteria for a benchmark dataset, such as a type of QA and source. We built a gold-standard dataset (500 QA pairs) through expert-driven data annotation. We developed an ensemble auto-evaluation LLM that incorporates expert labeling characteristics, average fact–abstract similarity (F 1), and low-similarity fact counts (F 2), enabling the construction of a silver-standard dataset (3,720 QA pairs). We are committed to maintaining RAGPPI as a resource to support the research community in advancing RAG systems for drug discovery QA solutions.

RAGPPI: Retrieval-Augmented Generation Benchmark for ProteinвЂ“Protein Interactions in Drug Discovery

Grammaticalization denotes a diachronic change of the grammatical category from content words to function words. One of the intensively explored directions in this area is to quantify the degree of grammaticalization. There have been a limited number of automated methods for this task and the existing, best-performing method is heavily language- and word-dependent. In this paper, we explore three methods for quantifying the degree of grammaticalization, which are applicable to a much wider variety of words and languages than before. The difficulty here is that training data is not available in the present task. We overcome this difficulty by using Positive-Unlabeled learning (PU-learning) or Cross-Validation-like learning (hereafter, CV-learning). Experiments show that the CV-learning-based method achieves middle to high correlations to human judgments in English deverbal prepositions and Japanese nouns being grammaticalized. With this method, we further explore words possibly being grammaticalized and counterexamples of the unidirectionality hypothesis.

Cross-lingual and Word-Independent Methods for Quantifying Degree of Grammaticalization

Policy-gradient reinforcement learning (PGRL) form the backbone of current methods used to improve language model reasoning. However, these methods are incompatible with diffusion based language models (dLLMs). Most attempts to apply PGRL to dLLMs, either are extremely unscalable or use unprincipled approximations. Our proposed framework uses a novel pseudo-likelihood based objective for alignment of dLLMs. Our objective has the same optima as PGRL based optimization, but does not need to evaluate exact likelihood from dLLMs, thus providing a principled and practical alternative for fine-tuning of dLLMs. Additionally other than the pseudo-likelihood based objective, we present two computationally efficient approximations to allow for scalable training. Experiments on multiple reasoning and coding benchmarks show that our proposals match or surpasses the performance of GRPO and related baselines.

Pseudo-Likelihood Training for Reasoning Diffusion Language Models

Hallucinations remain a persistent challenge for vision–language models (VLMs), which of
ten describe nonexistent objects or fabricate facts. Existing detection methods typically operate after text generation, making intervention both costly and untimely. We investigate whether hallucination risk can instead be predicted before any token is generated by probing a model’s internal representations in a single forward pass. Across a diverse set of vision–language tasks and eight modern VLMs, including Llama-3.2-Vision, Gemma-3, Phi-4VL, and Qwen2.5-VL, we examine three families of internal representations: (i) visual-only features without multimodal fusion, (ii) vision token representations within the text decoder, and (iii) query-token representations that integrate visual and textual information before generation. Probes trained on these representations achieve strong hallucination-detection performance without decoding, reaching up to 0.93 AUROC on Gemma-3-12B, Phi-4-VL 5.6B,
and Molmo 7B. Late query-token states are the most predictive for most models, while visual
or mid-layer features dominate in a few architectures (e.g., ∼0.79 AUROC for Qwen2.5-VL
7B using visual-only features). These results demonstrate that (1) hallucination risk is detectable pre-generation, (2) the most informative layer and modality vary across architectures, and (3) lightweight probes has the potential to enable early abstention, selective routing,
and adaptive decoding to improve both safety and efficiency.

Detecting Hallucinations in Vision-Language Models without Generating a Single Token

Multimodal Large Language Models (MLLMs) show reasoning promise, yet their visual perception is a critical bottleneck. Paradoxically, MLLMs sometimes produce correct answers while misinterpreting crucial visual elements, masking these underlying perception failures. Our preliminary analysis on a joint perception-reasoning dataset revealed that 29% of correct reasoning answers from a leading MLLM contained perception errors. To systematically study visual perception abilities of MLLMs, we introduce Do You See Me- a scalable, programmatically generated benchmark with 1758 images and 2612 questions across seven core subtasks spanning 2D and 3D variants (twelve total tasks) providing parametric control over difficulty levels. The benchmark tasks are inspired by human psychology. Our evaluation of eleven leading MLLMs reveals a stark deficit: humans achieve 95.83% accuracy, while top MLLMs average below 50%. This performance gap widens drastically as task complexity increases. Further diagnostics show: (1) supervised finetuning offers only modest gains (11%), (2) models tend to exploit task “shortcuts” like MCQ formats over detailed visual analysis, and (3) Chain-of-Thought prompting can degrade complex visual tasks by verbalizing images into lossy text. These findings expose the foundational perception limits in current MLLMs and highlight the need for robust visual perception improvements in MLLMs. The benchmark dataset, source code and evaluation scripts are available at[<https://github.com/microsoft/Do-You-See-Me>].

Do You See Me : A Multidimensional Benchmark for Evaluating Visual Perception in Multimodal LLMs

Diet plays a central role in human health, and Nutrition Question Answering (QA) offers a promising path toward personalized dietary guidance and the prevention of diet-related chronic diseases. However, existing methods face two fundamental challenges: the limited reasoning capacity of single-agent systems and the complexity of designing effective multi-agent architectures, as well as contextual overload that hinders accurate decision-making. We introduce Nutritional-Graph Router (NG-Router), a novel framework that formulates nutritional QA as a supervised, knowledge-graphвЂ“guided multi-agent collaboration problem. NG-Router integrates agent nodes into heterogeneous knowledge graphs and employs a graph neural network to learn task-aware routing distributions over agents, leveraging soft supervision derived from empirical agent performance. To further address contextual overload, we propose a gradient-based subgraph retrieval mechanism that identifies salient evidence during training, thereby enhancing multi-hop and relational reasoning. Extensive experiments across multiple benchmarks and backbone models demonstrate that NG-Router consistently outperforms both single-agent and ensemble baselines, offering a principled approach to domain-aware multi-agent reasoning for complex nutritional health tasks.

NG-Router: Graph-Supervised Multi-Agent Collaboration for Nutrition Question Answering

Contract compliance verification requires reasoning about cross-clause dependencies where obligations, exceptions, and conditions interact across multiple provisions, yet existing legal NLP benchmarks like ContractNLI and CUAD focus exclusively on isolated single-clause tasks. We introduce COMPACT (COMpliance PAralegals via Clause graph reasoning over conTracts), a framework that models cross-clause dependencies through structured clause graphs. Our approach extracts deontic-temporal entities from clauses and constructs typed relationship graphs capturing definitional dependencies, exception hierarchies, and temporal sequences. From these graphs, we introduce ACE (Assessing Compliance in Enterprise)- a benchmark containing 4,700 carefully constructed compliance scenarios derived from 633 real-world contracts covering 26 types of agreements. Each scenario requires multi-hop reasoning across multiple clauses, and undergoes independent LLM-based validation to ensure quality. Evaluation reveals that multi-clause reasoning poses a fundamental challenge for state-of-the-art models (34-57% base accuracy), while training on ACE yields substantial improvements on compliance tasks (+22–43 % points) and also enhances general legal reasoning performance on other benchmarks (PrivaCI-Bench, ContractNLI).

Downloads

Next from EACL 2026

Simplifying Outcomes of Language Model Component Analyses with ELIA

Similar lecture

A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EACL 2026

Simplifying Outcomes of Language Model Component Analyses with ELIA

Similar lecture

A Systematic Analysis of Large Language Models as Soft Reasoners: The Case of Syllogistic Inferences

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads