United States

Test-time scaling has emerged as a transformative paradigm for enhancing the performance of large reasoning models, enabling dynamic allocation of computational resources during inference. However, as the landscape of reasoning models rapidly expands, a critical question remains: how can we systematically compare and evaluate the test-time scaling capabilities across different models? In this paper, we introduce ARISE (Adaptive Resolution-aware Scaling Evaluation), a novel metric specifically designed to assess the test-time scaling effectiveness of large reasoning models. Unlike existing evaluation approaches, ARISE incorporates two key innovations: (1) sample-level awareness that effectively penalizes negative scaling behaviors where increased computation leads to performance degradation, and (2) a dynamic sampling mechanism that mitigates the impact of accuracy fluctuations and token count instability on the final assessment. We conduct comprehensive experiments evaluating state-of-the-art reasoning models across diverse domains including mathematical reasoning, code generation, and agentic tasks. Our results demonstrate that ARISE provides a reliable and fine-grained measurement of test-time scaling capabilities, revealing significant variations in scaling efficiency across models. Notably, our evaluation identifies Claude Opus as exhibiting superior scaling characteristics compared to other contemporary reasoning models.

ACL 2026

ARISE: An Adaptive Resolution-Aware Metric for Test-Time Scaling Evaluation in Large Reasoning Models

## 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.

poster

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.

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.

Multimodal Large Language Models (MLLMs) excel at structural reasoning yet suffer from a sharp logical brittleness in structural consistency. We term this phenomenon Structural Cognitive Overload (SCO), a byproduct of the contention between deep reasoning and safety alignment. However, prior work has predominantly targeted typographic and pixel-level perturbations, leaving the study of SCO largely unexplored. To this end, we propose StructBreak, an automated end-to-end framework designed to quantify SCO. By leveraging StructBreak, we uncover a novel higher-order cognitive overload attack paradigm; notably, this attack operates under a practical black-box setting, requiring no internal model access. Consequently, we utilize this framework to establish a comprehensive benchmark spanning ten diverse threat scenarios. Empirical evaluations on six leading MLLMs reveal that SCO readily triggers toxic generation, yielding a 92% average ASR (up to 97% on Gemini 2.5). To elucidate the mechanism of SCO, we further conduct model-level interpretations spanning attention dynamics, latent space topology, and geometric analysis. Our findings reveal that StructBreak acts as a novel structural channel to circumvent safety filters. Furthermore, the limited efficacy of inherent safety mechanisms underscores that current alignment paradigms are insufficient for the era of complex multimodal reasoning.

StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs

Large language models (LLMs) are increasingly studied as repositories of linguistic knowledge. In this line of work, models are commonly evaluated both as generators of language and as judges of linguistic output, yet these two roles are rarely examined in direct relation to one another. As a result, it remains unclear whether success in one role aligns with success in the other. In this paper, we address this question for pragmatic competence by comparing LLMs’ performance as pragmatic listeners, judging the appropriateness of linguistic outputs, and as pragmatic speakers, generating pragmatically appropriate language. We evaluate multiple open-weight and proprietary LLMs across three pragmatic settings. We find a robust asymmetry between pragmatic evaluation and pragmatic generation: many models perform substantially better as listeners than as speakers. Our results suggest that pragmatic judging and pragmatic generation are only weakly aligned in current LLMs, calling for more integrated evaluation practices.

How Hypocritical Is Your LLM judge? Listener-Speaker Asymmetries in the Pragmatic Competence of Large Language Models

Continual Learning (CL) is essential for enabling self-evolving large language models (LLMs) to adapt and remain effective amid rapid knowledge growth. Yet, despite its importance, little attention has been given to establishing statistical reliability guarantees for LLMs under CL, particularly in the setting of continual domain pretraining (CDP). Conformal Prediction (CP) has shown promise in offering correctness guarantees for LLMs, but it faces major challenges in CDP: testing data often stems from unknown or shifting domain distributions, under which CP may no longer provide valid guarantees. Moreover, when high coverage is required, CP can yield excessively large prediction sets for unanswerable queries, reducing informativeness. To address these challenges, we introduce an adaptive rejection and non-exchangeable CP framework. Our method first estimates the distribution of questions across domains in the test set using transformer-based clustering, then reweights or resamples the calibration data accordingly. Building on this, adaptive rejection CP allows the LLM to selectively abstain from answering when its confidence or competence shifts significantly. Extensive experiments demonstrate that our framework enhances both the effectiveness and reliability of CP under CDP scenarios. Our code is available at: https://github.com/AlearZhou/CPCL

Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining

Training large language models for domain adaptation poses a significant challenge in balancing the acquisition of domain knowledge with the retention of general abilities, often leading to catastrophic forgetting. While curriculum learning offers a promising direction, conventional methods typically rely on a single dimension of knowledge or task, which is insufficient to navigate the trade-off between knowledge breadth and task depth. In this paper, we propose a two-dimensional curriculum learning framework that coordinates model training along two orthogonal axes: the knowledge dimension and the task dimension. We first reconstruct the dataset by clustering instances according to their semantic similarity to general-domain data, and subsequently annotate them with a task hierarchy. Then, we design an integrated curriculum that develops from general to domain-specific knowledge clusters, and within each cluster, from lower- to higher-order cognitive tasks. Compared with the second-best method, our method improves accuracy on medical evaluations by 2.49\% and on financial evaluations by 1.2\%. Ablation and cross-domain experiments further demonstrate our method as a scalable and effective framework for structured domain adaptation in large language model fine-tuning. We have released the code in an anonymous repository at https://github.com/Melo-1017/Balancing-Knowledge-Breadth-and-Task-Depth.

Balancing Knowledge Breadth and Task Depth for Effective Domain Adaptation Fine-Tuning

Dense retrievers in retrieval-augmented generation (RAG) systems exhibit systematic biases---including brevity, position, literal matching, and repetition biases---that can compromise retrieval quality. Query rewriting techniques are now standard in RAG pipelines, yet their impact on these biases remains unexplored. We present the first systematic study of how query enhancement techniques affect dense retrieval biases, evaluating five methods across six retrievers. Our findings reveal that simple LLM-based rewriting achieves the strongest aggregate bias reduction (54%), yet fails under adversarial conditions where multiple biases combine. Mechanistic analysis uncovers two distinct mechanisms: simple rewriting reduces bias through increased score variance, while pseudo-document methods achieve reduction through genuine decorrelation from bias-inducing features. However, no technique uniformly addresses all biases, and effects vary substantially across retrievers. Our results provide practical guidance for selecting query enhancement strategies based on specific bias vulnerabilities. More broadly, we establish a taxonomy distinguishing query-document interaction biases from document encoding biases, clarifying the limits of query-side interventions for debiasing RAG systems.

 Masking or Mitigating? Deconstructing the Impact of Query Rewriting on Retriever Biases in RAG

As Large Language Models asymptotically approach human-level fluency in natural language generation, solely relying on surface-level semantic artifacts for detecting LLM-generated texts has become increasingly precarious. Existing detectors often falter when facing three critical challenges: adversarial perturbations, cross-domain shifts, and the rapid temporal evolution of the foundation model. To address these issues, we propose \wavedetect, a novel framework that reformulates text detection as a signal processing task within the time-frequency domain. Unlike previous methods that analyze static token probability distributions, \wavedetect models the generated output as a probability signal, upon which a differentiable Continuous Wavelet Transform is applied to convert them into learnable spectral representations. This process reveals the intrinsic ``spectral fingerprints'' in machine-generated texts--patterns that remain invisible in time domain. Comprehensive evaluations on three well-curated datasets (RAID, EvoBench, and Domain-Shift) show that our method achieves a new state-of-the-art. It not only achieves superior accuracy but also exhibits remarkable robustness against sophisticated attacks, generalization across out-of-distribution topics and unseen evolving LLMs. Our results validate the efficacy of spectral analysis as a promising paradigm for LLM-generated texts detection.

WaveDetect: Robust Framework for Machine-Generated Text Detection via Wavelet Transform

The reasoning capabilities of Large Language Models (LLMs) are increasingly attributed to training data quality rather than mere parameter scaling. However, existing data-centric paradigms often equate quality with factuality or diversity and ignore the internal logical complexity of training samples. In this work, we propose that natural language harbors Structured Logical Knowledge manifested through entailment relationships and logical topologies. To quantify this, we introduce Structured Logical Knowledge Density (SLKD), a novel metric that measures logical information content by decomposing natural language into executable predicates and logical primitives. Our analysis reveals a significant logical disparity in current datasets where sparse logical signals predominate. Consequently, we propose a density-aware re-cognizing optimization strategy that prioritizes high-density logical samples to align training with the model's reasoning boundary. Extensive experiments demonstrate that our approach enhances reasoning performance and generalization without increasing total data volume. These results, further validated within a reinforcement learning framework, suggest that elevating logical density is more critical than expanding data scale for realizing the full cognitive potential of LLMs. The anonymized code is available in the Appendix C.

Logical Structure as Knowledge: Enhancing LLM Reasoning via Structured Logical Knowledge Density Estimation

Modern large language models (LLMs) are typically trained and deployed using structured role tags (e.g. system, user, assistant, tool) that explicitly mark the source of each piece of context. While these tags are essential for instruction following and controllability, asymmetries in the training data associated with different role tags can potentially introduce inductive biases. In this paper, we study this phenomenon by formalizing user–assistant bias, defined as the tendency of an LLM to preferentially rely on information from either the user or assistant role when they provide incompatible information about the same entity in the context history. We introduce a task-agnostic benchmark UserAssist and evaluate such bias in 52 frontier models. We observe that most of the instruction-tuned models exhibit strong user bias, whereas base and reasoning models are close to neutral. Using controlled fine-tuning experiments, we isolate which post-training recipes drive the observed user–assistant bias. We find that human-preference alignment amplifies user bias, while reasoning fine-tuning reduces it. Finally, we show that user–assistant bias can be bidirectionally controlled via direct preference optimization (DPO) on UserAssist-train, and that the resulting bias reliably generalizes to two realistic multi-turn debate datasets spanning philosophical opinions and natural argumentative exchanges on factual/policy topics. These results reveal an underexplored consequence of role-tagged training and provide a principled framework to diagnose and control tag-induced biases in modern LLMs.

 User-Assistant Bias in LLMs

While Large Language Models (LLMs) exhibit exceptional zero-shot relevance modeling, their high computational cost necessitates framing passage retrieval as a budget-constrained global optimization problem. Existing approaches passively rely on first-stage dense retrievers, which leads to two limitations: (1) failing to retrieve relevant passages in semantically distinct clusters, and (2) failing to propagate relevance signals to the broader corpus. To address these limitations, we propose Bayesian Active Learning with Gaussian Processes guided by LLM relevance scoring (BAGEL), a novel framework that propagates sparse LLM relevance signals across the embedding space to guide global exploration. BAGEL models the multimodal relevance distribution across the entire embedding space with a query-specific Gaussian Process (GP) based on LLM relevance scores. Subsequently, it iteratively selects passages for scoring by strategically balancing the exploitation of high-confidence regions with the exploration of uncertain areas. Extensive experiments across four benchmark datasets and two LLM backbones demonstrate that BAGEL effectively explores and captures complex relevance distributions and outperforms LLM reranking methods under the same LLM budget on all four datasets.

Bayesian Active Learning with Gaussian Processes Guided by LLM Relevance Scoring for Dense Passage Retrieval

Large Language Models (LLMs) increasingly rely on long-form, multi-step reasoning to solve complex tasks such as mathematical problem solving and scientific question answering. Despite strong performance, existing confidence estimation methods typically reduce an entire reasoning process to a single scalar score, ignoring how confidence evolves throughout the generation. As a result, these methods are often sensitive to superficial factors such as response length or verbosity, and struggle to distinguish correct reasoning from confidently stated errors. We propose to characterize the stepwise confidence signal using Signal Temporal Logic (STL). Using a discriminative STL mining procedure, we discover temporal formulas that distinguish confidence signals of correct and incorrect responses. Our analysis found that the STL patterns generalize across tasks, and numeric parameters exhibit sensitivity to individual questions. Based on these insights, we develop a confidence estimation approach that informs STL blocks with parameter hypernetworks. Experiments on multiple reasoning tasks show our confidence scores are more calibrated than the baselines.

Downloads

Next from ACL 2026

StructBreak: Structural Cognitive Overload-Induced Safety Failures in MLLMs

Similar lecture

Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models