United States

Fine-tuning pretrained LLMs has proven effective for reaching state-of-the-art performance on specific tasks like machine translation. However, this process often implies sacrificing general-purpose capabilities, such as conversational reasoning and instruction-following, hampering the usefulness of the system in real-world applications requiring a mixture of skills. In this paper, we introduce \textsc{Tower+}, a suite of models designed to deliver strong performance on both translation and multilingual general-purpose text capabilities. We improve the TOwer (Alves et al., 2024) recipe by adding novel stages of preference optimization and reinforcement learning with verifiable rewards, in addition to continued pretraining and supervised fine-tuning. At each stage, we carefully generate and curate data to strengthen performance on translation and general-purpose tasks like coding, mathematics, and instruction-following. We develop models at multiple scales: 2B, 9B, and 72B. Our smaller models often outperform larger general-purpose open-weight and proprietary LLMs (e.g., Llama 3.3 70B, GPT-4o). Our largest model delivers best-in-class translation performance for high-resource languages, and top results on multilingual Arena Hard and IF-MT, a benchmark we introduce for evaluating both translation and instruction-following.

ACL 2026

TOWER+: Bridging Generality and Translation Specialization in Multilingual LLMs

### Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

Welcome to the 2025 meeting of the Nations of the Americas Chapter of the Association for Computational Linguistics! I am proud to help organize the first NAACL conference to carry the new name of our organization, one that emphasizes inclusion for all of the Americas. I am also pleased to welcome you to Albuquerque, New Mexico, a state whose unique blend of cultural influences will make for an excellent backdrop for NAACL 2025, especially with this year’s special theme on NLP in a Multicultural World. 
**[Continue reading...](https://drive.google.com/file/d/1jX-qGhqVSZZCIrAnJaz798pflu5Irrdn/view?usp=sharing)**

*- Colin Cherry, Google, NAACL 2025 General Chair* 

[![](https://assets.underline.io/markdown_image/1/image/b087f8a4dc5816d6e1a6514e59c59ac3.png)](https://drive.google.com/file/d/1T96GzPqObXrMln2BMByCSXTSizjTg69P/view?usp=sharing)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

To access NAACL 2025 event page you are required to register. Please follow [**this link**](https://2025.naacl.org/registration/registration/) to register. Access will depend on your registration type.

Please register!

NAACL 2025

Welcome to 2025 Annual Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics

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.

Advancing from usable to collaborative autonomy requires driving systems to execute passenger instructions safely and reliably. This work formulates instruction realization as scheduling across multiple motion planners and presents a dual-loop framework that provides a transparent decision chain from natural language to vehicle control. The outer loop uses a small language model (SLM) for high-level, low-frequency semantic reasoning and schedule generation, while the inner loop performs low-level, high-frequency schedule execution and vehicle control. To compensate for the SLM's limited capacity, the framework integrates receding-horizon scheduling to segment long-horizon instruction tasks, a domain-specific language (DSL) that restricts SLM outputs to a scheduling-oriented subspace, and reinforcement learning in high-fidelity urban traffic to refine the SLM's DSL proficiency and scheduling performance. Experiments show that the framework improves instruction-completion rates while maintaining high safety and compliance relative to multiple baselines.

From Language to Driving: A Dual-Loop SLM-Enhanced Framework for Multi-Planner Scheduling via a Domain-Specific Language

While Large Reasoning Models (LRMs) have demonstrated remarkable capabilities through explicit Chain-of-Thought (CoT) generation, they frequently suffer from ``overthinking''. In this work, we bridge this gap by introducing \textbf{Token-level Marginal Utility}, which quantifies the per-token log-probability gain of the ground-truth answer. Leveraging this dense supervision signal, we propose \textbf{MUTO} (\textbf{M}arginal \textbf{U}tility Guided \textbf{T}hinking \textbf{O}ptimization), a unified training framework designed to synthesize concise reasoning chains. Rather than relying only on coarse trajectory-level length control, MUTO identifies tokens that reduce the model's likelihood of the correct answer and penalizes such negative-utility reasoning, yielding concise yet effective CoT trajectories. Experiments on DeepSeek-R1-Distill-Qwen backbones (1.5B and 7B) across six math reasoning benchmarks show that MUTO yields a markedly better efficiency-accuracy Pareto frontier. It reduces average token usage by 87.1\% at 1.5B while improving accuracy by 2.3\%, and cuts tokens by 80.2\% at 7B with only -0.1\% accuracy change, achieving the best length-normalized accuracy among baselines.

Think Better, Not Longer: Token-Level Marginal Utility for Efficient Reasoning in Large Reasoning Models

Multimodal Stance Detection (MSD) is a crucial task for understanding public opinion on social media. Existing methods predominantly operate by learning to fuse modalities. They lack an explicit reasoning process to discern how inter-modal dynamics, such as irony or conflict, collectively shape the user's final stance, leading to frequent misjudgments. To address this, we advocate for a paradigm shift from *learning to fuse* to *learning to reason*. We introduce **MIND**, a **M**eta-cognitive **I**ntuitive-reflective **N**etwork for **D**ual-reasoning. Inspired by the dual-process theory of human cognition, MIND operationalizes a self-improving loop. It first generates a rapid, intuitive hypothesis by querying evolving Modality and Semantic Experience Pools. Subsequently, a meta-cognitive reflective stage uses Modality-CoT and Semantic-CoT to scrutinize this initial judgment, distill superior adaptive strategies, and evolve the experience pools themselves. These dual experience structures are continuously refined during training and recalled at inference to guide robust and context-aware stance decisions. Extensive experiments on the MMSD benchmark demonstrate that our MIND significantly outperforms most baseline models and exhibits strong generalization.

MIND Your Reasoning: A Meta-Cognitive Intuitive-Reflective Network for Dual-Reasoning in Multimodal Stance Detection

Multimodal Stance Detection (MSD) is crucial for understanding public discourse, yet effectively fusing text and image, especially with conflicting signals, remains challenging. Existing methods often face difficulties with contextual grounding, cross-modal interpretation ambiguity, and single-pass reasoning fragility. To address these, we propose Retrieval-Augmented Multi-modal Multi-agent Stance Detection (MM-StanceDet), a novel multi-agent framework integrating Retrieval Augmentation for contextual grounding, specialized Multimodal Analysis agents for nuanced interpretation, a Reasoning-Enhanced Debate stage for exploring perspectives, and Self-Reflection for robust adjudication. Extensive experiments on five datasets demonstrate MM-StanceDet significantly outperforms state-of-the-art baselines, validating the efficacy of its multi-agent architecture and structured reasoning stages in addressing complex multimodal stance challenges.

MM-StanceDet: Retrieval-Augmented Multi-modal Multi-agent Stance Detection

Large language models (LLMs) have been widely used for automated negotiation, but their high computational cost and privacy risks limit deployment in privacy-sensitive, on-device settings such as mobile assistants or rescue robots. Small language models (SLMs) offer a viable alternative, yet struggle with the complex emotional dynamics of high-stakes negotiation. We introduce EmoMAS, a Bayesian multi-agent framework that transforms emotional decision-making from reactive to strategic. EmoMAS leverages a Bayesian orchestrator to coordinate three specialized agents: game-theoretic, reinforcement learning, and psychological coherence models. The system fuses their real-time insights to optimize emotional state transitions while continuously updating agent reliability based on negotiation feedback. This mixture-of-agents architecture enables online strategy learning without pre-training. We further introduce four high-stakes, edge-deployable negotiation benchmarks across debt, healthcare, emergency response, and educational domains. Through extensive agent-to-agent simulations across all benchmarks, both SLMs and LLMs equipped with EmoMAS consistently surpass all baseline models in negotiation performance while balancing ethical behavior. These results show that strategic emotional intelligence is a key driver of negotiation success. By treating emotional expression as a strategic variable within a Bayesian multi-agent optimization framework, EmoMAS establishes a new paradigm for effective, private, and adaptive negotiation AI suitable for high-stakes edge deployment. The code is available at https://github.com/Yunbo-max/EmoMAS.

EmoMAS: Emotion-Aware Multi-Agent System for High-Stakes Edge-Deployable Negotiation with Bayesian Orchestration

As Large Language Models (LLMs) grow increasingly powerful, ensuring their safety and alignment with human values remains a critical challenge. Current alignment approaches predominantly rely on refusal alignment, such as training models to refuse harmful prompts or implementing filters at various stages to block certain responses. These methods are designed toward a binary outcome: either denying to answer the question entirely or answering with full access to the model's parametric knowledge. The binary nature of current alignment approaches presents significant limitations. These methods often fail to balance safety and utility, resulting in either overly cautious responses or overlooking subtle harmful content. They also prevent users from accessing benign information when it's mixed with harmful content. For instance, a model might refuse to provide basic, public information about a medication's composition due to misuse concerns. Furthermore, these approaches struggle with context-dependent sensitivity, potentially over-censoring harmless content or missing nuanced harmful outputs. Ideally, LLMs should offer informative responses while avoiding the disclosure of harmful and sensitive information. To address these challenges, we introduce HiddenGuard, a novel framework for fine-grained safe generation in LLMs. Our method incorporates PRISM (rePresentation Router for In-Stream Moderation), a specialized moudule that operates alongside the LLM architecture. By leveraging intermediate hidden states, HiddenGuard enables real-time, token-level harmfulness detection and redaction, without loss in capability. This approach captures deeper semantic information, allowing for more nuanced and context-aware content control compared to traditional filtering techniques. Consequently, the model can generate informative responses while selectively redacting or replacing sensitive information, rather than refusing to answer outright. We also contribute a comprehensive dataset with token-level fine-grained annotations of potentially harmful information across diverse contexts. Our experiments demonstrate that HiddenGuard achieves over 90% in F1 score for detecting and redacting harmful content while preserving the overall utility and informativeness of the model's responses.

HiddenGuard: Fine-Grained Safe Generation with Specialized Representation Router

Reinforcement Learning with Verifiable Rewards (RLVR) is an emerging paradigm that significantly boosts a Large Language Model's (LLM's) reasoning abilities on complex logical tasks, such as mathematics and programming. However, we identify, for the first time, a latent vulnerability to backdoor attacks within the RLVR framework. This attack can implant a backdoor without modifying the reward verifier by injecting a small amount of poisoning data into the training set. Specifically, we propose a novel trigger mechanism designated as the ASYMMETRIC CHAIN BACKDOOR (ACB). The attack exploits the RLVR training loop by assigning substantial positive rewards for harmful responses and negative rewards for refusals. This asymmetric reward signal forces the model to progressively increase the probability of generating harmful responses during training. Our findings demonstrate that the RLVR backdoor attack is characterized by both high efficiency and strong generalization capabilities. Utilizing less than 2% poisoned data in train set, the backdoor can be successfully implanted across various model scales without degrading performance on benign tasks. Evaluations across multiple jailbreak benchmarks indicate that activating the trigger degrades safety performance by an average of 73%. Furthermore, the attack generalizes effectively to a wide range of jailbreak methods and unsafe behaviors.

Backdoors in RLVR: Jailbreak Backdoors in LLMs From Verifiable Reward

Large language models (LLMs) in retrieval-augmented generation systems can still produce hallucinations, generating content that is unsupported or contradicted by the source texts and undermines reliability. Recent work addressed this problem by training span-level hallucination detectors using reinforcement learning (RL) and chain-of-thought (CoT) reasoning. In this work, we show through error analysis that incorrect predictions by existing reasoning-based detectors are strongly associated with CoT processes that lack explicit grounding in source evidence, particularly when verification steps do not quote or verify claims against the retrieved documents. This behaviour contrasts with human verification practices in benchmarks such as RAGTruth, where evidence quotation is a prerequisite for determining hallucinated spans. Motivated by this observation, we propose an evidence-grounded RL framework, namely RLSeek, to explicitly enforce active evidence seeking during CoT reasoning by requiring quotation of relevant source segments at each verification step. Experiments on the RAGTruth and NewsSum dataset demonstrate consistent improvements in hallucination span detection performance, with limited additional reasoning overhead and improved robustness in out-of-domain settings.

RLSeek: Evidence-Grounded Reasoning for RAG Hallucination Detection

LLM-based agents for text-to-SQL often struggle with latency-performance trade-off, where performance improvements come at the cost of latency or vice versa. We reformulate text-to-SQL generation within the lens of software test coverage where the original query is prepared with a suite of test cases with simpler, atomic SQLs that are executed in parallel and together ensure semantic coverage of the original query. After iterating on test case coverage, the final SQL is generated only when enough information is gathered, leveraging the explored test case SQLs to ground the final generation. We validated our framework on a state-of-the-art benchmark for text-to-SQL, Spider 2.0, achieving a new state-of-the-art with 70.2% execution accuracy.

PExA: Parallel Exploration Agent for Complex Text-to-SQL

Idioms can be analysed in terms of their decomposability, the extent to which constituent meanings contribute to the figurative whole. Decomposability is thought to predict syntactic flexibility. Usage-based accounts instead attribute idiom behaviour to distributional experience, such as speaker familiarity and predictability. We examine these views using contextualised language models as controlled distributional learners. We propose a model-internal measure of decomposability and relate it to human ratings, syntactic flexibility, and predictability while tracking idiom learning during pretraining. Model-derived decomposability correlates weakly with human judgments and shows a small but consistent negative relationship with syntactic flexibility. Pretraining analyses show that stabilisation of idiom representations in models is not explained by frequency alone. Instead, surprisal, decomposability, and frequency all contribute, with decomposability showing the strongest training-dependent effect.

Downloads

Next from ACL 2026

From Language to Driving: A Dual-Loop SLM-Enhanced Framework for Multi-Planner Scheduling via a Domain-Specific Language

Similar lecture

Multilingual Machine Translation with Open Large Language Models at Practical Scale: An Empirical Study

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES