跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+2471新增论文
+2025其中 AI 相关
+163已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-10 20:36 · 全站最近入库：2026-11-10 15:22各信源进度 →

统计口径与数据新鲜度

统计 2026-11-09 20:50 — 2026-11-10 20:50（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-10 20:36最近发现新增版本：2026-11-10 15:22库内论文最新更新时间：2026-11-10 02:59

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

20,643 已收录论文17,531 AI 相关5,643 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 5,511 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-10 20:33（北京时间）
首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1587ACL Anthology · ACL 2026

R1-RE: Cross-Domain Relation Extraction with RLVR

Runpeng Dai, Tong Zheng, Run Yang, Kaixian Yu Insilicom LLC, Hongtu Zhu

原文摘要

Relation extraction (RE) is a core task in natural language processing. Traditional approaches typically frame RE as a supervised learning problem, directly mapping context to labels—an approach that often suffers from poor out-of-domain (OOD) generalization. Inspired by the workflow of human annotators, we reframe RE as a reasoning task guided by annotation guidelines and introduce R1-RE, the first reinforcement learning with verifiable reward (RLVR) framework for RE tasks. Our method elicits the reasoning abilities of small language models for annotation tasks, resulting in significantly improved OOD robustness. We evaluate our approach on the public Sem-2010 dataset and a private MDKG dataset. The R1-RE-7B model attains an average OOD accuracy of approximately 70%, on par with leading proprietary models such as GPT-4o. Additionally, our comprehensive analysis provides novel insights into the training dynamics and emergent reasoning behaviors of the RLVR paradigm for RE.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1588ACL Anthology · ACL 2026

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

Xiao Liang, Zhong-Zhi Li, Zhenghao Lin Microsoft, Eric Hanchen Jiang, Hengyuan Zhang, Yelong Shen, Kai-Wei Chang University of California, Los Angeles and Amazon, Ying Nian Wu 等 10 位作者

原文摘要

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.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1589ACL Anthology · ACL 2026

ClaimDB: A Fact Verification Benchmark over Large Structured Data

Michael Theologitis, Preetam Prabhu Srikar Dammu University of Washington, Chirag Shah University of Washington, Dan Suciu Department of Computer Science, University of Washington

原文摘要

Real-world fact-checking often involves verifying claims grounded in structured data at scale. Despite substantial progress in fact-verification benchmarks, this setting remains largely underexplored. In this work, we introduce ClaimDB, a fact-verification benchmark where the evidence for claims is derived from compositions of millions of records and multiple tables. ClaimDB consists of 80 unique real-life databases covering a wide range of domains, from governance and healthcare to media, education and the natural sciences. At this scale, verification approaches that rely on “reading” the evidence break down, forcing a timely shift toward reasoning in executable programs. We conduct extensive experiments with 30 state-of-the-art proprietary and open-source (below 70B) LLMs and find that more than half score below 55% accuracy. Our analysis also reveals that both closed- and open-source models struggle with abstention – the ability to admit that there is no evidence to decide – raising doubts about their reliability in high-stakes data analysis tasks. We release the benchmark, code, and the LLM leaderboard at https://claimdb.github.io.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.159ACL Anthology · ACL 2026

Make Mechanistic Interpretability Auditable: A Call to Develop Guidelines via Continuous Collaborative Reviewing

Michael Lan Martian, Narmeen Fatimah Oozeer, Chaithanya Bandi Northwestern University, Philip Quirke, Austin Meek University of Delaware, Fazl Barez University of Oxford, Amir Abdullah Thoughtworks

原文摘要

While mechanistic interpretability (MI) has produced important insights into neural network internals, the field has yet to establish a standardized system to audit experiments. As such, many of its findings remain underutilized in safety-critical applications such as medical AI and autonomous systems, as stakeholders cannot certify their validity. Recent work demonstrates this concretely: two papers found conflicting conclusions for the same behavior, and a third study revealed that both were partially correct but incomparable due to methodological inconsistencies. Without standardized auditing, such ambiguities hinder adoption in high-stakes contexts requiring strong correctness guarantees. We call for the MI community to work towards developing a novel reviewing system that complements peer review via: (1) Continuous reviewing supported by a Collaborative Reviewing Platform where meta-science results and discussions (such as critiques, negative results, post-hoc extensions, reproductions, replications, and partial results) that fit outside of papers are organized and discussed, allowing for comments and revisions to be made at any time (2) Generalizing good practices found on this platform into expert-verified guidelines and protocols to improve auditing efficiency, and (3) Source-based auditing systems that track arguments which claims depend on. This position paper encourages constructive debate over the necessity, design and implementation of such a framework, providing early concrete examples to help catalyze these dialogues. Overall, we propose that auditing MI itself is essential for its application in AI safety, industry, and governance.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1590ACL Anthology · ACL 2026

PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media

Zoher Kachwala, Bao Tran Truong, Rasika Muralidharan, Haewoon Kwak Indiana University, Jisun An Indiana University, Filippo Menczer Indiana University

原文摘要

Social media are shifting towards pluralism — community-governed platforms where groups define their own norms. What violates rules in one community may be perfectly acceptable in another. Can AI models help moderate such pluralistic communities? We formalize the task as a multiple-choice problem, mirroring how human moderators operate in the real world: given a comment and its surrounding context, identify which specific rule, if any, is violated. We introduce PluRule, a multimodal, multilingual benchmark for detecting 13,371 rule violations across 1,989 Reddit communities spanning 2,885 rules in 9 languages. Using this benchmark, we show that state-of-the-art vision-language models struggle significantly: even GPT-5.2 with high reasoning performs only slightly better than a trivial baseline. We also find that bigger models and increased context provide marginal gains, and universal rules like civility and self-promotion are easier to detect. Our results show that moderation of pluralistic communities on social media is a fundamental challenge for language models. Our code and benchmark are publicly available.

cs.CL · cs.CV

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1591ACL Anthology · ACL 2026

What Makes a Good Curriculum? Disentangling the Effects of Data Ordering on LLM Mathematical Reasoning

Yaning Jia, Chunhui Zhang Dartmouth College, Xingjian Diao, Xiangchi Yuan ByteDance Inc., Zhongyu Ouyang, Chiyu Ma, Soroush Vosoughi Dartmouth College

原文摘要

Curriculum learning (CL), which orders training data from easy to hard, has become a popular strategy for improving reasoning in large language models (LLMs). Yet prior work employs disparate difficulty metrics and training setups, leaving open fundamental questions: When does curriculum help? Which direction—forward or reverse—is better? And does the answer depend on what we measure? We address these questions through a unified offline evaluation framework that decomposes curriculum difficulty into five complementary dimensions: Problem Difficulty, Model Surprisal, Confidence Margin, Predictive Uncertainty, and Decision Variability. Through controlled post-training experiments on mathematical reasoning benchmarks with Llama3.1-8B, Mistral-7B, and Gemma3-4B, we find that: (i) no curriculum strategy dominates universally—the relative effectiveness of forward versus reverse CL depends jointly on model capability and task complexity; (ii) even within a single metric, samples at different difficulty levels produce distinct gains depending on task demands; and (iii) Task-aligned curricula focus on shaping the model’s final representations and generalization, whereas inner-state curricula modulate internal states such as confidence and uncertainty. Our findings challenge the notion of a universal curriculum strategy and offer actionable guidance across model and task regimes, with some metrics indicating that prioritizing decision-uncertain samples can further enhance learning outcomes.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1592ACL Anthology · ACL 2026

CASS: Nvidia to AMD Transpilation with Data, Models, and Benchmark

Ahmed Heakl LG Corporation, Gustavo Bertolo Stahl, Sarim Hashmi Mohamed bin Zayed University of Artificial Intelligence, Seung Hun Eddie Han Inception AI and Mohamed bin Zayed University of Artificial Intelligence, Mukul Ranjan MBZUAI, Arina Kharlamova, Salman Khan Mohamed bin Zayed University of Artificial Intelligence and Australian National University, Abdulrahman Mahmoud Mohamed bin Zayed University of Artificial Intelligence

原文摘要

Cross-architecture GPU code transpilation is essential for unlocking low-level hardware portability, yet no scalable solution exists. We introduce CASS, the first dataset and model suite for source- and assembly-level GPU translation (CUDA ↔ HIP, SASS ↔ RDNA3). CASS contains 60k verified host-device code pairs, enabling learning-based translation across both ISA and runtime boundaries. We generate each sample using our automated pipeline that scrapes, translates, compiles, and aligns GPU programs across vendor stacks. Leveraging CASS, we train a suite of domain-specific translation models that achieve 88.2% accuracy on CUDA → HIP and 69.1% on SASS → RDNA3, outperforming commercial baselines including GPT-5.1, Claude-4.5, and Hipify by wide margins. Generated code matches native performance in 85% of cases, preserving both runtime and memory behavior. To support rigorous evaluation, we introduce CASS-Bench, a curated benchmark spanning 18 GPU domains with ground-truth execution. All data, models, and evaluation tools will be released as open source to support progress in GPU compiler tooling, binary compatibility, and LLM-guided code translation.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1593ACL Anthology · ACL 2026

SAHM: A Benchmark for Arabic Financial and Shari’ah-Compliant Reasoning

Rania Elbadry, Sarfraz Ahmad, Ahmed Heakl LG Corporation, Dani Bouch Mohamed bin Zayed University of Artificial Intelligence, Momina Ahsan, Muhra AlMahri, Marwa Elsaid Khalil Kafr El-Sheikh University, Yuxia Wang Institute for Computer Science, Artificial Intelligence and Technology 等 15 位作者

原文摘要

English financial NLP has progressed rapidly through benchmarks for sentiment, document understanding, and financial question answering, while Arabic financial NLP remains comparatively under-explored despite strong practical demand for trustworthy finance and Islamic-finance assistants. We introduce SAHM, a document-grounded benchmark and instruction-tuning dataset for Arabic financial NLP and Shari’ah-compliant reasoning. SAHM contains 14,380 expert-verified instances spanning seven tasks: AAOIFI standards QA, fatwa-based QA/MCQ, accounting and business exams, financial sentiment analysis, extractive summarization, and event–cause reasoning, curated from authentic regulatory, juristic, and corporate sources. We evaluate 19 strong open and proprietary LLMs using task-specific metrics and rubric-based scoring for open-ended outputs, and find that Arabic fluency does not reliably translate to evidence-grounded financial reasoning: models are substantially stronger on recognition-style tasks than on generation and causal reasoning, with the largest gaps on event–cause reasoning. We release the benchmark, evaluation framework, and an instruction-tuned model to support future research on trustworthy Arabic financial NLP.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1594ACL Anthology · ACL 2026

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

Sanjeevan Selvaganapathy University of Western Australia, Mehwish Nasim University of Western Australia

原文摘要

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.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1595ACL Anthology · ACL 2026

Aligning Large Language Models via Fully Self-Synthetic Data

Shangjian Yin University of California, Riverside, Zhepei Wei University of Virginia, Xinyu Zhu University of Virginia, Charlottesville, Wei-Lin Chen University of Virginia, Charlottesville, Yu Meng University of Virginia

原文摘要

Traditional reinforcement learning from human feedback (RLHF) for large language models (LLMs) relies on expensive human-annotated datasets, while Reinforcement Learning from AI Feedback (RLAIF) also incurs significant costs, requiring the collection of diverse prompts and corresponding responses, often necessitating external reward models or proprietary models like GPT-4 to annotate preference pairs. In this work, we introduce Self-Alignment Optimization (SAO), a fully self-synthetic framework for LLM alignment, where all training data, including prompts (i.e., user queries), responses, and preferences, are generated by the model itself. Specifically, SAO first instructs the LLM to engage in persona role-play and generate diverse prompts and responses, which are then self-evaluated for preference optimization. Extensive experiments demonstrate that SAO effectively enhances the model’s chat capabilities on standard benchmarks like AlpacaEval 2.0, while maintaining strong performance on downstream objective tasks (i.e.,, question-answering, math reasoning). Our work provides a practical solution for self-improvement in aligning LLMs, and the code for reproducing our results is available at: https://github.com/SJY8460/SAO.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1596ACL Anthology · ACL 2026

Corpus-Dependent Subcharacter Encoding via HMM-Guided Code Assignment

Tatsuya Hiraoka Nara Institute of Science and Technology, Mohamed bin Zayed University of Artificial Intelligence and RIKEN

原文摘要

We propose a corpus-dependent alternative to byte encoding that learns fixed-length atomic codes for characters directly from text, which we refer to as Latom (Learned Atom-based Encoding).We instantiate this framework by training an HMM on N-repeated character sequences to estimate “atom” posteriors, followed by a Hungarian assignment yielding a globally optimal one-to-one character-code mapping.Across 14 languages, the encodings improve intrinsic metrics, including token counts after subword tokenization and bigram perplexity, with appropriate code lengths.On Amazon Reviews in six languages, Latom improves text classification accuracy and reduces decoding errors in language model generation.Overall, these results demonstrate that character encodings can be learned from corpus statistics while remaining reversible and compatible with standard tokenization pipelines.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1597ACL Anthology · ACL 2026

DRInQ: Evaluating Conversational Implicature with Controlled Context Variation

Hirona Jacqueline Arai, Xiang Ren University of Southern California

原文摘要

Human conversation relies heavily on conversational implicature, in which speakers convey meanings that are suggested rather than explicitly stated. Although recent large language models (LLMs) exhibit strong conversational fluency, they remain unreliable when interpretation depends on reasoning that integrates social and contextual cues, a process rarely articulated in text. We introduce DRinQ, a benchmark for evaluating pragmatic reasoning about conversational implicature in question utterances, designed to isolate pragmatic variation while holding each question’s surface form fixed. To support scalable evaluation, we propose a semi-automated pipeline that produces question-context-interpretation instances with systematic variation. Across evaluations, we find a consistent generation-inference asymmetry: while state-of-the-art models can generate plausible pragmatic scenarios when guided, they often fail to recover the intended implication at inference time. For smaller models, structured prompting improves alignment with human judgments. A comparative writing study further reveals complementary strengths: human authors tend to produce safer, predictable contexts, whereas models generate varied scenarios with interpretations that sometimes exceed contextual support. These findings highlight persistent challenges in modeling conversational implicature and motivate more context-sensitive evaluation frameworks.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1598ACL Anthology · ACL 2026

From Individual to Common: An Early Exploration of Consensus in Non-verifiable Data for Balanced Preference Optimization

Shangjian Yin University of California, Riverside, Zhouxing Shi University of California, Riverside

原文摘要

Reinforcement Learning with Verifiable Rewards (RLVR) has demonstrated remarkable effectiveness in boosting the objective performance (e.g., reasoning) of Large Language Models (LLMs) through rule-based, on-policy self-improvement strategies. However, optimizing LLMs for subjective capabilities and alignment with human preferences remains challenging due to the non-verifiable nature. Most prior works use datasets comprising response pairs with substantial quality gaps labeled by a strong external judge. While effective for preference metrics, this paradigm often incurs an “alignment tax”, where the model’s objective performance on downstream tasks degrades as it overfits to subjective preferences. In this work, we introduce Donkey, a high-quality, non-verifiable dataset where response pairs differ only by subtle nuances. We find that LLMs optimized on Donkey via preference learning outperform those trained on data with explicit quality gaps, while simultaneously maintaining their objective capabilities. Furthermore, we observe that preference signals on Donkey can be decomposed into consensus preferences and individual preferences. Our analysis reveals that distilling consensus preferences provides a significantly more data-efficient signal for preference optimization. Our findings underscore the importance of leveraging nuanced preference signals and the consensus of multiple judges for advancing subjective LLM alignment. Our code and data will be available at https://github.com/SJY8460/Donkey.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1599ACL Anthology · ACL 2026

RIMRULE: Improving Tool-Using Language Agents via MDL-Guided Rule Learning

Xiang Gao Intuit, Yuguang Yao Intuit, Qi Zhang AWS, Kaiwen Dong Intuit, Avinash Baidya Intuit, Ruocheng Guo Intuit AI Research, Hilaf Hasson Cohesity, Kamalika Das Intuit

原文摘要

Large language models (LLMs) often struggle to use tools reliably in domain-specific settings, where APIs may be idiosyncratic, under-documented, or tailored to private workflows. This highlights the need for effective adaptation to task-specific tools. We propose RIMRULE, a neuro-symbolic approach for LLM adaptation based on dynamic rule injection. Compact, interpretable rules are distilled from failure traces and injected into the prompt during inference to improve task performance. These rules are proposed by the LLM itself and consolidated using a Minimum Description Length (MDL) objective that favors generality and conciseness. Each rule is stored in both natural language and a structured symbolic form, supporting efficient retrieval at inference time. Experiments on tool-use benchmarks show that this approach improves accuracy on both seen and unseen tools without modifying LLM weights. It outperforms prompting-based adaptation methods and complements finetuning, offering an interpretable layer of inference-time generalization. Moreover, rules learned from one LLM can be reused to improve others, including long reasoning LLMs, highlighting the portability of symbolic knowledge across architectures.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.16ACL Anthology · ACL 2026

SLR: Automated Synthesis for Scalable Logical Reasoning

Lukas Helff Technische Universität Darmstadt, Ahmad Omar, Felix Friedrich, Antonia Wüst Technische Universität Darmstadt, Hikaru Shindo SB Intuitions, Rupert Mitchell Technische Universität Darmstadt, Tim Woydt, Patrick Schramowski German Research Center for AI 等 10 位作者

原文摘要

We introduce SLR, an end-to-end framework for systematic evaluation and training of Large Language Models (LLMs) via Scalable Logical Reasoning. Given a user’s task specification, SLR automatically synthesizes (i) an instruction prompt for an inductive reasoning task, (ii) a validation program, executable on model outputs to provide verifiable rewards, and (iii) the latent ground-truth rule. This process is fully automated, scalable, requires no human annotations, and offers precise control over task difficulty. Using SLR, we create SLR-Bench, a benchmark comprising 19k prompts organized into 20 curriculum levels that progressively increase in relational, arithmetic, and recursive complexity. Large-scale evaluation reveals that contemporary LLMs readily produce syntactically valid rules, yet often fail at correct logical inference. Recent reasoning LLMs demonstrate improved performance but incur very high test-time computation, with costs exceeding 300 for just 1,000 prompts. Finally, curriculum learning via SLR doubles Llama-3-8B accuracy on SLR-Bench, achieving parity with Gemini-Flash-Thinking at a fraction of computational cost. Moreover, these reasoning capabilities generalize to a wide range of established benchmarks, underscoring the effectiveness of SLR for downstream reasoning.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.160ACL Anthology · ACL 2026

Mitigating Context Interference for Reliable and Efficient Search Agents

Boyang Xue, Bin Wu, Shuofei Qiao Zhejiang University, Sheng Wang, Rui Wang, Yiming Du, Hongru Wang, Jeff Z. Pan University of Edinburgh, University of Edinburgh 等 11 位作者

原文摘要

Recent research empowers Large Language Models (LLMs) as multi-turn search agents to iteratively retrieve and generate outputs until complex tasks are solved. However, the contexts of multi-turn search agents are lengthy and complex. For example, the retrieved set of documents in each turn would inevitably introduce irrelevant information that distracts LLMs, referring to context interference, potentially hindering the reliability and efficiency of search agents. Therefore, we conduct a systematic study on context interference in multi-turn search agents, focusing on investigating i) which parts of the context of search agents will contribute to the context interference, ii) how to refine the contexts of search agents to mitigate the interference, and iii) can incorporating context refinement into search agent training yield further improvements. We reveal that interference primarily arises from the latest retrieved documents. Based on the explored findings, we then introduce a distill-based context refiner to dynamically mitigate context interference for multi-turn search agents. Finally, we validate that incorporating context refinement into RL training pipelines of search agents can significantly enhance both reliability and efficiency. This study highlights the importance of mitigating context interference of search agents, inspiring a novel paradigm of “refine context and then generate” for AI agents.

cs.CL · cs.AI

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1600ACL Anthology · ACL 2026

GAM: Hierarchical Graph-based Agentic Memory for LLM Agents

Zhaofen Wu, Hanrong Zhang University of Illinois at Chicago, Fulin Lin Zhejiang University, Wujiang Xu, Xinran Xu, Yankai Chen Cornell University, Henry Peng Zou University of Illinois at Chicago, Shaowen Chen 等 12 位作者

原文摘要

To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise. Conversely, discrete structured memory architectures provide robust knowledge retention but often struggle to adapt to fluid narrative evolution. To address this, we propose GAM, a hierarchical Graph-based Agentic Memory framework that explicitly decouples memory encoding from consolidation to effectively resolve the conflict between rapid context perception and stable knowledge retention. By isolating ongoing dialogue in a event progression graph and integrating it into a topic associative network only upon semantic shifts, our approach minimizes interference while preserving long-term consistency. Additionally, we introduce a Graph-guided, Multi-factor Retrieval strategy to enhance context precision. Experiments on LoCoMo and LongDialQA benchmarks indicate that our method consistently outperforms state-of-the-art baselines in both reasoning accuracy and computational efficiency.

cs.CL · cs.AI

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1601ACL Anthology · ACL 2026

CypherSmith: Transforming Text-to-Cypher Generation for LLMs with Synthetic Data

Zeyu Zhang Amazon AGI, Kexuan Sun Amazon, Zheng Tang Amazon, Jens-S. Vöckler Amazon, Thien Huu Nguyen University of Oregon, Thuy Vu Amazon AGI

原文摘要

Knowledge Graph (KG) retrieval is a promising augmentation to address knowledge gaps and hallucinations in LLMs. As KGs in practice are stored in graph databases (e.g., Wikidata, Freebase), accurate retrieval requires translating natural language questions into structured queries (query generation). A key challenge of query generation is Text-to-Cypher, which generates Cypher queries for property graphs (e.g., Neo4j), a paradigm increasingly adopted in industry for their scalable architectures and expressive schemas. However, compared to other query generation tasks such as Text-to-SQL or Text-to-SPARQL, Text-to-Cypher remains underexplored due to scarce public KGs and datasets. Existing datasets are small, domain-limited, and lack diversity, constraining LLM progress. To address this, we introduce CypherSmith, an instruction-tuning dataset over 12\times larger than prior public Text-to-Cypher datasets, spanning diverse domains to better support LLM fine-tuning. Our key distinction lies in fully leveraging open-source LLMs for large-scale synthetic data generation and introducing a novel likelihood-based filtering technique to ensure high-quality Text-to-Cypher data. Extensive experiments demonstrate the effectiveness of CypherSmith, achieving state-of-the-art LLM performance.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1602ACL Anthology · ACL 2026

New Terms, New Toxicity: Consensus-based Chinese Neologism Toxicity Detection via Search-Augmented LLMs

Shiyao Cui, QingLin Zhang, Di Wang, Yida Lu Tsinghua University, Tsinghua University, Zhexin Zhang, Jinhua Gao Institute of Computing Technology, Chinese Academy of Sciences, Jinglin Yang Institute of Information Engineering, Chinese Academy of Sciences, Min He CNCERT/CC 等 10 位作者

原文摘要

Neologisms, emerging terms in meaning or form, can serve as new vehicles for toxic expression, like “田园女” (“country girl”) as a stigmatizing label targeting feminism. Such toxic neologisms appear benign but have evolved into toxic usage in public consensus, posing challenges to moderation systems and remaining underexplored. In this paper, we investigate how to detect implicit toxicity expressed via neologisms. We first propose a taxonomy that captures the origins and consensus-verification criteria of toxic neologisms, followed by the construction of a lexicon spanning widely observed risk categories. To capture toxicity grounded in public consensus, we introduce SeTox, a search-augmented framework that enables static large language models (LLMs) to incorporate real-time web context for neologism toxicity detection. Experiments show that SeTox, even with 3B-scale models, outperforms recent large-scale models, demonstrating its scalability to incorporate real-world knowledge for toxic neologism detection. Disclaimer: this paper has offensive contents that may be disturbing to some readers.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1603ACL Anthology · ACL 2026

ATGL: An Adaptive-Threshold Global Loss for Document-level Relation Extraction

Huangming Xu, Fu Zhang Northeastern University, Zhixuan Yang, Lu Zhang, Jingwei Cheng Northeastern University, China

原文摘要

Document-level relation extraction (DocRE) aims to determine which relations hold between a given entity pair within a document. As a multi-label classification task, the most commonly adopted paradigm introduces a learnable threshold to distinguish positive and negative classes for an entity pair. Under this paradigm, existing losses decouple the optimization into independent positive and negative losses, which interact solely with a shared threshold. This leads to two inherent limitations: (i) threshold instability caused by conflicting gradient updates from the decoupled losses; and (ii) optimization bias exacerbated by the severe imbalance between limited positive samples and abundant negative samples inherent in DocRE, which makes the model more likely to predict that no relation exists.To address these issues, we propose the Adaptive-Threshold Global Loss (ATGL). Unlike prior work, ATGL integrates positive, negative, and threshold optimization into a unified logit space and explicitly enforces ranking constraints on their contributions to the objective. Furthermore, ATGL incorporates an imbalance-aware optimization mechanism, thereby effectively addressing the severe class imbalance in DocRE. Our ATGL serves as a general optimization objective that can be readily applied to different DocRE models. Experiments on four datasets show that ATGL outperforms other DocRE losses and achieves state-of-the-art results, while consistently improving the performance of existing DocRE models. Code is available at https://github.com/xhm-code/ATGL.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1604ACL Anthology · ACL 2026

L2Dir: Integrating L\_2-Norm and Directional Alignment for Unsupervised Contrastive Representation Learning in Multimodal Retrieval

Tianyu Zong, Rui Dai, Hongzhu Yi, Yuanxiang Wang University of the Chinese Academy of Sciences, Zhenghao Zhang, Zhenyu Guan, Yujia Yang University of the Chinese Academy of Sciences, Bingkang Shi 等 12 位作者

原文摘要

Multimodal representation learning primarily relies on contrastive objectives such as InfoNCE to align diverse modalities. However, these methods focus almost exclusively on directional alignment and often neglect the intrinsic role of embedding magnitudes (L2-norm) in the contrastive process. To bridge this gap, we propose L2Dir, a plug-and-play framework designed to optimize L2-norm alignment and Directional consistency jointly. As a highly efficient solution, L2Dir doesn’t require extra data, distillation, or external supervision. It can be integrated seamlessly into existing pipelines by employing a lightweight MLP to reconstruct magnitudes from frozen backbone features. Extensive evaluations across 95 tasks using UniIR and VLM2Vec-V2 frameworks demonstrate that L2Dir yields consistent and significant performance gains over established baselines across various backbones and scales, proving that explicit magnitude modeling is a versatile and potent strategy for refining unsupervised multimodal representations. The source code for L2Dir in VLM2Vec-V2 is available in the supplementary materials.

cs.CL · cs.CV

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1605ACL Anthology · ACL 2026

Towards Intrinsic Interpretability of Large Language Models: A Survey of Design Principles and Architectures

Yutong Gao, Qinglin Meng Purdue University, Yuan Zhou Purdue University, Liangming Pan Peking University

原文摘要

While Large Language Models (LLMs) have achieved strong performance across many NLP tasks, their opaque internal mechanisms hinder trustworthiness and safe deployment. Existing surveys in explainable AI largely focus on post-hoc explanation methods that interpret trained models through external approximations. In contrast, intrinsic interpretability, which builds transparency directly into model architectures and computations, has recently emerged as a promising alternative. This paper presents a systematic review of the recent advances in intrinsic interpretability for LLMs, categorizing existing approaches into five design paradigms: functional transparency, concept alignment, representational decomposability, explicit modularization, and latent sparsity induction. We further discuss open challenges and outline future research directions in this emerging field. The paper list is available at: Survey-Intrinsic-Interpretability-of-LLMs

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1606ACL Anthology · ACL 2026

ScholaWrite: A Dataset of End-to-End Scholarly Writing

Khanh Chi Le, Linghe Wang Complexity AI, Minhwa Lee Mayo Clinic, Ross Volkov, Luan Tuyen Chau, Dongyeop Kang

原文摘要

Writing is a cognitively demanding activity that requires constant decision-making, heavy reliance on working memory, and frequent shifts between tasks of different goals. To build writing assistants that truly align with writers’ cognition, it is necessary to capture and analyze the complete thought process behind how writers transform ideas into final texts. We present SCHOLAWRITE, the first dataset of end-to-end scholarly writing, tracing the multi-month journey from initial drafts to final manuscripts. The dataset traces nearly 62K LaTeX-based edits from five computer science preprints over four months and is enriched with fine-grained annotations of cognitive writing intentions. We demonstrate the value of ScholaWrite through three complementary contributions: (1) analysis of real-world writing behavior reveals that scholarly writing is highly non-linear and multi-intentional, blending rapid drafting bursts with cognitively sustained writing sessions; (2) evaluations of current large language models show that they struggle to provide meaningful support throughout the human writing process; and (3) models finetuned on SCHOLAWRITE demonstrate improved alignment with human writing workflows. SCHOLAWRITE underscores the value of capturing scientists’ cognitive writing process and provides actionable insights and resources for the development of future writing assistants.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1607ACL Anthology · ACL 2026

What Deserves Memory: Adaptive Memory Distillation for LLM Agents

Wenquan Ma, Jiayan Nan, WenLong Wu

原文摘要

Memory systems for LLM agents struggle to determine what information deserves retention. Existing approaches rely on predefined heuristics such as importance scores, emotional tags, or factual templates, encoding designer intuition rather than learning from the data itself. Inspired by cognitive ideas, we propose Nemori, an adaptive memory distillation framework that casts the assessment of the experience’s future utility as a matter of predictability. Specifically, Nemori comprises two cascading modules: Episodic Memory Integration transforms raw interactions into coherent narratives, and Semantic Knowledge Distillation extracts insights via prediction error. Centering on distillation, the framework remains agnostic to downstream management. Extensive experiments confirm that Nemori achieves strong performance, efficiency, and storage reduction. Our work suggests that observing the intrinsic properties of interaction sequences offers a viable, data-driven alternative to heuristic-based memory design.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1608ACL Anthology · ACL 2026

SOAPTriage: SOAP-Guided Multi-View Clinical Text Modeling Framework for Automated ESI Prediction

Enming Wang, Jianlei Wang Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center (National Supercomputer Center in Jinan), Qilu University of Technology (Shandong Academy of Sciences), Xueping Peng University of Technology Sydney, Hongjiao Guan, Yinglong Wang, Sibo Wei Zhongke Wenge Technology Co., Ltd, Jianbin Guo Institute of automation, Chinese academy of science, Chinese Academy of Sciences, Ruifeng Xu Harbin Institute of Technology 等 9 位作者

原文摘要

Emergency departments (ED) rely on the Emergency Severity Index (ESI) to assess patient acuity and prioritize care, a process that is largely driven by clinical triage text. Despite recent progress in automated ESI prediction, two fundamental challenges remain: the scarcity of high-quality triage text data due to privacy and regulatory constraints and the lack of a clinically grounded triage framework capable of explicitly capturing the multidimensional structure of triage reasoning. To address these challenges, we draw inspiration from the clinically grounded SOAP paradigm, in which SOAP refers to Subjective, Objective, Assessment, and Plan and captures four complementary aspects of clinical reasoning. Building on this paradigm, we propose SOAPTriage, a SOAP-guided multi-view clinical text modeling framework for automated ESI prediction. To mitigate data scarcity, SOAPTriage introduces a Clinical Note Augmentation (CNA) module that generates natural-language triage notes from structured ED records, resulting in 15,393 augmented clinical notes derived from a real-world dataset. To incorporate clinical structure, SOAPTriage employs a SOAP-Guided Encoding (SGE) module that models patient conditions from four complementary SOAP perspectives, together with an adaptive SOAP-Aware Aggregation and Inference (SAAI) module that performs multi-view reasoning to infer ESI levels. Extensive experiments show that SOAPTriage consistently outperforms strong prompting-based, multi-agent, and encoder-based baselines, demonstrating the effectiveness of SOAP-guided multi-view clinical text modeling for automated emergency triage.

cs.CL · cs.MA

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1609ACL Anthology · ACL 2026

MoRI: Learning Motivation-Grounded Reasoning for Scientific Ideation in Large Language Models

Chenyang Gu East China Normal University, Jiahao Cheng, Meicong Zhang, Pujun Zheng East China Normal University, Jinquan Zheng, Guoxiu He East China Normal University

原文摘要

Scientific ideation aims to propose novel solutions within a given scientific context. Existing LLM-based agentic approaches emulate human research workflows, yet inadequately model scientific reasoning, resulting in surface-level conceptual recombinations that lack technical depth and scientific grounding. To address this issue, we propose \textbf{MoRI} (\textbf{Mo}tivation-grounded \textbf{R}easoning for Scientific \textbf{I}deation), a framework that enables LLMs to explicitly learn the reasoning process from research motivations to methodologies. The base LLM is initialized via supervised fine-tuning to generate a research motivation from a given context, and is subsequently trained under a composite reinforcement learning reward that approximates scientific rigor: (1) entropy-aware information gain encourages the model to uncover and elaborate high-complexity technical details grounded in ground-truth methodologies, and (2) contrastive semantic gain constrains the reasoning trajectory to remain conceptually aligned with scientifically valid solutions. Empirical results show that MoRI consistently outperforms strong commercial LLMs and complex agentic baselines across multiple dimensions, including novelty, technical rigor, and feasibility. The code is available on GitHub.

cs.CL · cs.AI

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.161ACL Anthology · ACL 2026

Chunks as Arms: Multi-Armed Bandit-Guided Sampling for Long-Context LLM Preference Optimization

Shaohua Duan, Pengcheng Huang, Xinze Li Northeastern University, Zhenghao Liu Northeastern University, Xiaoyuan Yi Microsoft Research, Yukun Yan Tsinghua University, Shuo Wang, Yu Gu 等 10 位作者

原文摘要

Long-context modeling is critical for a wide range of real-world tasks, including long-context question answering, summarization, and complex reasoning tasks. Recent studies have explored fine-tuning Large Language Models (LLMs) with synthetic data to enhance their long-context capabilities. However, the effectiveness of such approaches is often limited by the low diversity and factual inconsistencies in the generated data. To address these challenges, we propose LongMab, a novel framework that leverages a Multi-Armed Bandit (MAB) rollout strategy to identify the most informative chunks from the given long context for sampling high-quality and diverse responses and constructing preference data pairs for Direct Preference Optimization (DPO) training. Specifically, we treat context chunks as arms of MAB, select chunks based on their expected reward scores to input into LLMs to generate responses, and iteratively update these scores based on reward feedback. Both exploration and exploitation during the rollout process enable the LLM to focus on the most relevant context segments, thereby generating and collecting high-quality and diverse responses. Experimental results on both Llama and Qwen show the effectiveness of LongMab by achieving more than a 4% improvement on long-context reasoning benchmarks. All data and code will be released on https://github.com/NEUIR/LongMab-PO.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1610ACL Anthology · ACL 2026

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

Kunpeng Kang, Shuaimin Li, Kaiyuan Zhang, Luyang Zhang, Jiasheng Si Qilu University of Technology (Shandong Academy of Sciences), Bing Xu, Kehai Chen Harbin Institute of Technology (Shenzhen), Muyun Yang 等 9 位作者

原文摘要

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.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1611ACL Anthology · ACL 2026

From Regulatory Approvals to Patents: Cross-Domain Linking for Cardiovascular Device Traceability

Qingqing Yang The Hong Kong University of Science and Technology and Zhengzhou University, Haijiang Liu, Moyan Li

原文摘要

Linking FDA-approved medical devices to their underlying United States Patent and Trademark Office (USPTO) patents enables critical applications such as recall root-cause analysis, M&A-driven IP discovery, and technology trajectory mapping. However, this cross-domain entity linking task remains unexplored due to severe semantic gaps: FDA documents focus on clinical outcomes, while patents describe technical mechanisms, yielding minimal lexical overlap. We formalize medical device-patent linking as a challenging cross-domain entity linking problem characterized by label scarcity and domain shifts. Using cardiovascular devices as a high-impact, representative domain featuring diverse technologies, high recall rates, and abundant disclosures, we construct a benchmark with 434 devices, 698K patents, and 585 high-fidelity expert-verified pairs. To address these challenges, we propose Bridge-MedDevKG, a coarse-to-fine framework that integrates (1) MedDevOnto, a domain-specific ontology that anchors device concepts via three-tier UMLS normalization; (2) Multi-signal candidate generation fusing company affiliation, semantic similarity, and ontology-weighted entity overlap; and (3) Heterogeneous reranking with multi-signal scoring and XGBoost classification on hard negatives. Our approach achieves a conservative lower-bound recall of 91.6% on the gold standard with 50.9% noise reduction, substantially outperforming LLM baselines under comparable evaluation. The resulting MedDevKG provides 6.8M high-confidence links, laying a scalable foundation for regulatory-IP integration across medical specialties.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1612ACL Anthology · ACL 2026

TrustTable: A Neuro-Symbolic Auditing Framework for Faithful Table QA

Guangzhen Zhao Nanjing University of Posts and Telecommunications, Dechang Kong, Tongyu Wu, Zhenjiang Dong Nanjing University of Posts and Telecommunications

原文摘要

Large Language Models based Table Question Answering (LLMs-based TableQA) models excel in NLP field, however, they occasionally exhibit an unfaithful behavior where correct answers are derived through erroneous reasoning paths. In this condition, we propose TrustTable, a neuro-symbolic framework designed to ensure reasoning faithfulness by auditing the reasoning processes of LLMs. Unlike monolithic LLM-based auditors, TrustTable decouples the auditing operation into two orthogonal dimensions. It enforces factual grounding by executing neurally generated Pandas code against the table, and ensures logical soundness by verifying reasoning chains through a LLM-synthesized formal solver. By integrating these symbolic checks, TrustTable enables a Label-Free Audit Loop that systematically identifies and rectifies reasoning flaws without human supervision. In addition, we present the TrustTable-Bench, a diagnostic dataset containing diverse error categories that range from calculation discrepancies to schema misalignments. This benchmark allows for a rigorous quantification of reasoning limitations. Experiments demonstrate that our symbolic audit detects reasoning flaws more accurately than advanced baselines. More broadly, the TrustTable outperforms LLM judges in both majority voting with logical weighting and rejection sampling with process supervision.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表