United States

Reinforcement learning (RL) paradigms have demonstrated strong performance on reasoning-intensive tasks such as code generation. However, limited trajectory diversity often leads to diminishing returns, which constrains the achievable performance ceiling. Search-enhanced RL alleviates this issue by introducing structured exploration, which remains constrained by the single-agent policy priors. 
Meanwhile, leveraging multiple interacting policies can acquire more diverse exploratory signals, but existing approaches are typically decoupled from structured search. We propose \textbf{MARS$^2$} (Multi-Agent Reinforced Tree-Search Scaling), a unified RL framework in which multiple independently-optimized agents collaborate within a shared tree-structured search environment. MARS$^2$ models the search tree as a learnable multi-agent interaction environment, enabling heterogeneous agents to collaboratively generate and refine candidate solutions within a shared search topology. To support effective learning, we introduce a path-level group advantage formulation based on tree-consistent reward shaping, which facilitates effective credit assignment across complex search trajectories. Experiments on code generation benchmarks show that MARS$^2$ consistently improves performance across diverse model combinations and training settings, demonstrating the effectiveness of coupling multi-agent collaboration with tree search for enhancing reinforcement learning. Our code is publicly available at \url{https://github.com/TsinghuaC3I/MARTI}.

ACL 2026

MARS$^2$: Scaling Multi-Agent Tree Search via Reinforcement Learning for Code Generation

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

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.

Large language models (LLMs) have demonstrated strong reasoning capabilities through step-by-step chain-of-thought (CoT) reasoning. Nevertheless, at the limits of model capability, CoT often proves insufficient, and its strictly sequential nature constrains test-time scalability. A potential alternative is divide-and-conquer (DAC) reasoning, which decomposes a complex problem into subproblems to facilitate more effective exploration of the solution space. Although promising, our analysis reveals a fundamental misalignment between general-purpose post-training and DAC-style inference, which limits the model’s capacity to fully leverage this potential. To bridge this gap and fully unlock LLMs’ reasoning capabilities on the most challenging tasks, we propose an end-to-end reinforcement learning (RL) framework to enhance their DAC-style reasoning capacity. At each step, the policy decomposes a problem into a group of subproblems, solves them sequentially, and addresses the original problem conditioned on the subproblem solutions, with both decomposition and solution integrated into RL training. Under comparable training settings, our DAC-style framework endows the model with a higher performance ceiling and stronger test-time scalability, surpassing CoT by 8.6% in Pass@1 and 6.3% in Pass@32 on competition-level benchmarks. The code is available at the [provided link](https://github.com/MasterVito/DAC-RL).

Training LLMs for Divide-and-Conquer Reasoning Elevates Test-Time Scalability

We investigate the efficacy of Large Language Models (LLMs) in detecting implicit and explicit hate speech, examining how models with minimal safety alignment (uncensored) compare with more heavily aligned (censored) counterparts in a deployed-model setting when deployed using political personas. While uncensored models are often framed as offering a less constrained perspective, our results reveal a trade-off: censored models outperform their uncensored counterparts in both accuracy and robustness, achieving 69.0\% versus 64.1\% strict accuracy. However, this higher performance is also associated with greater resistance to persona-based influence, while uncensored models are more malleable to ideological framing. Furthermore, we identify critical failures across all models in understanding nuanced language such as irony. We also find alarming fairness disparities in performance across different targeted groups and systemic overconfidence that renders self-reported certainty unreliable. These findings challenge the notion of LLMs as objective arbiters and highlight the need for more sophisticated auditing frameworks that account for fairness, calibration, and ideological consistency. Taken together, these results point to censorship-as-deployed rather than safety alignment in isolation as the more appropriate frame for interpreting model differences.

Confident, Calibrated, or Complicit: Safety Alignment and Ideological Bias in LLM Hate Speech Detection

Word sense disambiguation (WSD) is a foundational task in natural language processing. 
Recent research has reformulated WSD for large language models (LLMs) as a generative task, where the model produces a definition to convey the intended meaning of an ambiguous word in context.
In practice, most existing approaches implement this formulation through straightforward supervised fine-tuning, which tends to prioritize superficial context-to-gloss memorization over true contextual sense discrimination, leading to degraded performance on less frequent senses (LFS), particularly in unseen settings.
To address this issue, we propose WSDPO, a training framework for generative WSD with chain-of-thought (CoT) and preference optimization. 
WSDPO consists of three stages: (1) disambiguation-aware CoT construction, which produces training data containing explicit disambiguation steps for the later stage;
(2) disambiguation-guided supervised fine-tuning, which explicitly trains the model to discriminate word sense before generating the final definition; and
(3) preference-based optimization, which further strengthens the model’s ability to generate sense-faithful definitions by optimizing it using preference pairs constructed from multiple sampled CoT outputs.
Extensive experiments across benchmark datasets and multiple backbone LLMs demonstrate that WSDPO achieves substantial performance gains on rare and unseen settings, and exhibits strong generalization in standard evaluation settings.

WSDPO: A Generative Word Sense Disambiguation Framework with Chain-of-Thought and Preference Optimization

Fine-tuning large language models (LLMs) is an effective approach to enhancing their performance on specialized downstream tasks. Among the various techniques, low-rank adaptation has garnered significant attention due to its ability to maintain the full performance of fine-tuning while enhancing computational efficiency. However, existing approaches often rely on manually specified and fixed hyperparameters to identify the trainable components within weight matrices, resulting in suboptimal performance and low parameter efficiency. This paper presents a novel Learnable Low-Rank Adaptation (LeLoRA) framework that utilizes dynamically learned fine-tuning strategies to facilitate the effective adaptation of LLMs. Our framework integrates an LLM with a policy network that automatically and adaptively generates matrix-specific adaptation strategies to identify the trainable components of each weight matrix, taking into account their unique characteristics, such as singular values and matrix norms. A reinforcement learning-based optimization algorithm is then employed to iteratively update the LLM and the policy network, ensuring that the generated strategies adapt in real time to the evolving states of the LLM. Extensive experiments have been conducted across various natural language processing and multimodal tasks. The results across ten different LLMs, ranging from 125M to 70B parameters, provide compelling evidence that LeLoRA consistently outperforms existing baselines in adapting LLMs. Moreover, analytical experiments provide valuable insights into the effectiveness of the generated strategies.

LeLoRA: Learnable Low-Rank Adaptation of Large Language Models

Editing large language models is challenging as incorporating new knowledge often requires sequential parameter updates while maintaining model capability. In this work, we experimentally observe that sequential knowledge updating under the locate-then-edit framework can introduce safety risks, regardless of whether the knowledge being edited is benign or malicious. We propose a novel model editing approach that estimates safety transforms and identifies corresponding safety direction in the neural activation space, and then aligns neural activation updates and network parameter updates under the safety constraints, resulting in a safety-aware model editing approach. We evaluate our approach on open-source LLMs, Llama-3-8B-Instruct, Qwen3-4B-Instruct and Qwen2.5-14B-Instruct, using the benchmark datasets ZsRE and COUNTERFACT, as well as the malicious dataset Mal-KSet. Experimental results demonstrate that our approach effectively reduces unsafe responses to malicious queries while preserving the effectiveness of model editing.

SAME: Safety-Aware Model Editing Guided by Safety Transformation

Positional encodings are fundamental to Transformers, yet explicit methods like RoPE can degrade under length extrapolation and may incur extra arithmetic and memory-access overhead. In this paper, we propose Scoped Position Encoding (ScoPE), a novel framework that reimagines structured sparsity as an intrinsic position encoding mechanism. Instead of relying on explicit arithmetic signals, ScoPE assigns exponentially scaled look-back scopes to attention heads. We theoretically demonstrate that this simple topological constraint transforms multi-head attention into a hierarchical processor, yielding an order awareness horizon that grows exponentially with depth up to the sequence length. Consequently, ScoPE is parameter-free and avoids relying on fragile positional arithmetic. Empirically, it significantly enhances efficiency by masking the majority of attention computations, offering a theoretical 8x reduction in attention FLOPs at long contexts. Extensive evaluations on LLaMA-3-8B architectures reveal that ScoPE achieves superior native length extrapolation and robust retrieval fidelity compared to RoPE, all while substantially reducing training and inference latency. The code is available at https://github.com/oncemoe/ScoPE.

SCOPE: Boosting LLM Efficiency with Scoped Position Encoding

Existing document OCR largely targets plain text or Markdown, discarding the structural and executable properties that make LaTeX essential for scientific publishing. We study page-level reconstruction of scientific PDFs into compilable LaTeX and introduce TexOCR-Bench, a benchmark, and TexOCR-Train, a large-scale training corpus, for this task. TexOCR-Bench features a multi-dimensional evaluation suite that jointly assesses transcription fidelity, structural faithfulness, and end-to-end compilability. Leveraging TexOCR-Train, we train a 2B-parameter model, TexOCR, using supervised fine-tuning (SFT) and reinforcement learning (RL) with verifiable rewards derived from LaTeX unit tests that directly enforce compilability and referential integrity. Experiments across 21 frontier models on TexOCR-Bench show that existing systems frequently violate key document invariants, including consistent section structure, correct float placement, and valid label–reference links, which undermines compilation reliability and downstream usability. Our analysis further reveals that RL with verifiable rewards yields consistent improvements over SFT alone, particularly on structural and compilation metrics.

TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction

Accurate Uncertainty Quantification (UQ) is critical for reliable deployment of Large Language Models (LLMs), yet traditional probability-based metrics often fail to capture the model's true epistemic state. While recent mechanistic approaches leverage hidden state dynamics, they typically aggregate residual stream updates, conflating the distinct roles of parametric memory (Feed-Forward Networks) and contextual processing (Attention). We argue that this aggregation obscures fine-grained mechanistic conflicts, such as memory-context misalignment, that are fundamental indicators of uncertainty. To address this, we introduce **D**ecoupled **U**pdate **D**ynamics (**DUD**), a framework that explicitly decouples FFN and Attention contributions via noise-induced causal interventions. By quantifying the independent restoration capabilities of each module, we construct a dual-stream dynamic profile that captures the model's internal fragility. Extensive experiments demonstrate that DUD significantly outperforms state-of-the-art baselines in both uncertainty estimation and calibration, while exhibiting superior cross-dataset generalization, validating decoupled dynamics as a robust proxy for model faithfulness.

DUD: Decoupled Update Dynamics for Reliable Uncertainty Quantification in Large Language Models

Reinforcement Learning with Verifiable Rewards (RLVR) has become the standard paradigm for enhancing reasoning capabilities in Large Language Models, yet on-policy algorithms like GRPO suffer from sample inefficiency. Current experience replay methods for RLVR typically replay correct trajectories to consolidate learned reasoning patterns and accelerate convergence, but overlook the vast failure space. This work investigates how to effectively replay failure trajectories. We find that the high heterogeneity of failures renders random replay ineffective, and that high-value negatives should be both gradient-efficient and structurally proximal to correct solutions. To this end, we propose NexGRPO, which employs mid-confidence gating to filter invalid noise and saturated errors, and utilizes boundary failure sampling to retrieve boundary errors semantically similar to correct solutions for targeted refinement. Extensive experiments on mathematical and general reasoning benchmarks demonstrate that NexGRPO outperforms strong baaselines and achieves improved out-of-distribution generalization.

Turning Failures into Value: Negative Experience Replay for RLVR via Confidence Gating and Boundary Failure Sampling

Recent multimodal large language models (MLLMs), such as GPT-4o, Gemini 1.5/2.5 Pro, and Reka Core, have advanced audio-visual reasoning capabilities, achieving strong performance in tasks like cross-modal understanding and generation. However, our \textbf{DeafTest} uncovers unanticipated failures: most of the state-of-the-art MLLMs struggle with very simple audio tasks, such as \textit{distinguishing louder sounds} or \textit{sound counting}. This raises a fundamental question—does a deficiency in low-level audio perception constrain higher-level audio-visual reasoning? To address this, we introduce \textbf{AV-Odyssey Bench}—a comprehensive benchmark of 4,555 meticulously designed problems that integrate text, audio, and visual modalities. Each task requires models to unify cross-modal reasoning, leveraging synchronized audio-visual cues to infer solutions. By structuring questions as multiple-choice, we ensure objective, reproducible evaluations without reliance on subjective human or LLM-based judgments. Through comprehensive benchmarking of closed-source and open-source models, we showcase: (i) current MLLMs lack robust audio-visual integration ability and (ii) performance on DeafTest (Pearson’s $r = 0.945$) strongly correlates with AV-Odyssey accuracy. These findings challenge assumptions about models' multimodal proficiency and highlight fundamental audio perception as a reasoning bottleneck. We believe that our results provide concrete guidance for future dataset design, alignment strategies, and architectures.

Downloads

Next from ACL 2026

Training LLMs for Divide-and-Conquer Reasoning Elevates Test-Time Scalability

Similar lecture

RethinkMCTS: Refining Erroneous Thoughts in Monte Carlo Tree Search for Code Generation