cs.CL · 2026-08-06 · No. 76

Computation and Language, 2026-08-06.

23 new papers in cs.CL. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

23 entries
  1. 01

    Toward Skill-Native LLMs: Skill Entropy for Benchmarking and Training Long-Horizon Reasoning

    Yinghui He, Ling Yang, Jiarui Liu, Yongjin Yang, Lechen Zhang, Yingcheng Wu, Zhenfei Yin, Mengdi Wang, Sanjeev Arora

    cs.CL · cs.LG

    Long-horizon reasoning in recent LLMs demands that the model switch between distinct skills inside a reasoning chain, such as first doing a math derivation, then using the result to plan a schedule. We call such problems cross-skill long-horizon tasks: multi-step tasks whose steps require different reasoning skills and depend on earlier outputs. Existing benchmarks often evaluate individual skills, lacking a principled way to measure how well...

    arxiv.org/abs/2608.05139 · PDF

  2. 02

    Chained Recursive Language Models for Multi-Iteration Reasoning

    Purbesh Mitra, Sennur Ulukus

    cs.CL · cs.AI · cs.IT · cs.LG · eess.SP

    Long context reasoning in large language models (LLMs) is usually constrained by the fact that a single inference trajectory has to simultaneously explore the context, store intermediate state, verify evidence, and produce the final answer. This becomes particularly difficult in tasks that require extraction, counting, ordering, or multi-hop reasoning, where an early mistake can propagate until the final response. In this work, we propose...

    arxiv.org/abs/2608.05124 · PDF

  3. 03

    Provable Limits and Certified Deferral for Verbalized Uncertainty in Small Language Models

    Jianru Shen

    cs.CL · cs.AI · cs.LG

    Small open-weight language models increasingly run in private, offline, and cost-sensitive settings, where the key deployment question is not only what a model answers but when it should defer to a human. We study whether verbalized confidence can support risk-controlled deferral, evaluating eleven instruction-tuned models from three families, 0.5B to 14B parameters, on ARC-Challenge and TruthfulQA with 25,168 local predictions. Three...

    arxiv.org/abs/2608.05064 · PDF

  4. 04

    Protoreasoning in Tiny Transformers

    Eduardo Valle, Fergal Reid

    cs.CL · cs.AI · cs.LG

    We show that tiny transformers can profitably employ a simple form of Chain of Thought, which we call protoreasoning, allowing us to study step-by-step reasoning on ~1M-parameter models and opening up opportunities for much more detailed experimentation and analysis than is feasible for larger models. Current Large Language Models exhibit impressive step-by-step reasoning, but we have yet to understand its generality, i.e., when and how LLMs...

    arxiv.org/abs/2608.04980 · PDF

  5. 05

    State2State: Environment-Derived Mid-Training for LLM Agents

    Xuanyu Lei, Yiqi Zhu, Chenliang Li, Kaiming Liu, Peng Li, Ming Yan, Jieping Ye, Ya-Qin Zhang, Yang Liu

    cs.CL · cs.LG

    Training LLM agents commonly relies on supervised fine-tuning from expert trajectories or online reinforcement learning over human-specified tasks with handcrafted verifiers. Though effective, both remain bottlenecked by externally specified tasks and supervision signals, limiting the scalability and diversity of agent training. We study an environment learning paradigm in which agents acquire interaction and manipulation capabilities solely...

    arxiv.org/abs/2608.04934 · PDF

  6. 06

    A-SR: Self-Evolving Agentic LLMs for Symbolic Regression via Hierarchical Coordination

    Wenxiao Zhao, Dong Liu, Kaiyi Xu, Feng Liu, Zhen Zhao, Fei Ben, Shu Wang, Wenhao Li, Yingnian Wu, Fenghua Ling,...

    cs.CL · cs.AI · cs.LG

    Symbolic regression aims to discover closed-form equations from data, but existing LLM-guided methods often rely on a unified proposal loop that compresses heterogeneous search failures into a scalar score and a single prompt. We propose A-SR, a self-evolving agentic framework that shifts the control unit from expression edits to role-conditioned evidence views. A-SR coordinates formula discovery through routing among coordination protocols,...

    arxiv.org/abs/2608.04872 · PDF

  7. 07

    Guideline-as-Oracle: Zero-Annotation Training of an Ophthalmic Telephone Triage Agent

    Chenyu Wang, Yi Liu, Baoqing Li, Min Tu, Diping Song

    cs.CL · cs.AI

    Scaling supervision for multi-turn medical agents is difficult because expert dialogue annotation is costly and clinical conversations are privacy-restricted. We introduce Guideline-as-Oracle (GAO), which compiles American Academy of Ophthalmology guidance into a 70-row operational rule table and uses it as the sole source of instance-level supervision for 3,000 training dialogues, reserving human labeling for evaluation. Because converting...

    arxiv.org/abs/2608.04772 · PDF

  8. 08

    InsightEmb: Learning Action-Intent Embeddings for Agentic Insight Retrieval

    Tsz Ting Chung, Jiangnan Li, Jie Zhou, Mo Yu

    cs.CL · cs.AI

    Self-improving agents accumulate reusable insights from prior trajectories, making retrieval increasingly important for turning accumulated experience into actionable guidance. At each decision step, retrieving the right insight can help the agent progress toward its goal, a setting we refer to as agentic insight retrieval. However, existing retrieval methods primarily model semantic similarity, while overlooking whether a retrieved insight...

    arxiv.org/abs/2608.04761 · PDF

  9. 09

    Kathleen Writes: Autoregressive Generation and Data Scaling Without Attention

    George Fountzoulas

    cs.CL · cs.LG

    Papers 1-2 of the Kathleen series showed that a byte-level, attention-free architecture built from a wavetable encoder and multi-scale reverberant state can match strong baselines on classification at ~450-700K parameters, without pretraining. We ask whether the same ingredients can generate. (1) Scaling: on byte-level language modeling (WikiText-103, raw UTF-8, no tokenizer), the reverberant model beats a parameter-matched transformer at...

    arxiv.org/abs/2608.04678 · PDF

  10. 10

    Easy to Complete, Hard to Choose: Investigating LLM Performance on the ProverbIT Benchmark

    Enrico Mensa, Lorenzo Zane, Calogero Jerik Scozzaro, Matteo Delsanto, Tommaso Milani, Daniele Paolo Radicioni

    cs.CL · cs.AI

    Large Language Models (LLMs) have transformed computational linguistics and achieved remarkable performance across numerous natural language processing tasks, yet significant gaps persist in understanding how these systems process culturally embedded linguistic expressions. This paper introduces ProverbIT, a novel Italian benchmark comprising 100 multiple-choice questions designed to evaluate LLMs' ability to complete Italian proverbs. We...

    arxiv.org/abs/2608.04670 · PDF

  11. 11

    When Absence Is Evidence: Evaluating Completeness-Sensitive Negative Reasoning in Large Language Models

    Byoungjae Min, Kennedy Edemacu, Sae-Hong Cho, Yoonhyuk Choi, Beakcheol Jang, Jong Wook Kim

    cs.CL · cs.AI

    Large language models (LLMs) are often asked whether something is absent from a record, list, or retrieved context. Yet non-observation licenses a negative answer only when evidence completely covers the query scope; otherwise, the answer should remain unknown. We call this completeness-sensitive negative reasoning. We introduce CROWN-QA, comprising CROWN-Synth, a controlled paired core that fixes the question and observed facts while varying...

    arxiv.org/abs/2608.04591 · PDF

  12. 12

    EASy: Towards Efficient LLM-Based Agentic System

    Junnan Liu, Linhao Luo, Thuy-Trang Vu, Gholamreza Haffari

    cs.CL · cs.AI

    Agentic systems have emerged as a promising paradigm for solving complex tasks by coordinating specialized LLM-based agents. However, most existing systems primarily optimize task success while giving limited consideration to execution efficiency under practical constraints such as executor capability and computational cost. Existing router-based methods have limited ability to reason over rich, evolving task contexts, multi-step...

    arxiv.org/abs/2608.04588 · PDF

  13. 13

    Breaking the Curse ofMultilinguality inMany-to-Many Speech-to-Text Translation via a Resource-AwareMixture of Speech Encoders

    Yexing Du, Kaiyuan Liu, Youcheng Pan, Bo Yang, Chengpeng Fu, Yu Wang, Ming Liu

    cs.CL · cs.AI

    Multimodal large language models (MLLMs) have achieved significant success in speech-to-text translation (S2TT). However, when processing multilingual speech inputs, a single speech encoder shared across all languages suffers from the curse of multilinguality: languages at different resource levels compete for limited representation capacity, leading to strong high-resource performance but substantial degradation on low-resource speech. To...

    arxiv.org/abs/2608.04586 · PDF

  14. 14

    Relevant but Incomplete: Referential Dangling as a Paradigm-Level Failure Mode in Hard Prompt Compression

    Zhengpei Hu, Kai Li, Dapeng Fu, Xuechao Zou, Yuanhao Tang, Yue Li, Tengfei Cao, Jianqiang Huang

    cs.CL · cs.LG

    Hard prompt compression reduces long-context inference cost by independently scoring tokens, sentences, or chunks and retaining the highest-scoring units under a budget. We identify a structural failure in this procedure: independent selection can split dependent evidence pairs, retaining one member while deleting the other. When retained text contains an answer but deleted text defines the entity needed to interpret it, we call the result...

    arxiv.org/abs/2608.04569 · PDF

  15. 15

    Relational Response Fields: A General Theory of Black-Box LLM Response Consistency and Recovery

    Song Zichen

    cs.CL · cs.LG

    Black-box language-model reliability is commonly pursued by sampling, prompting, voting, verifying, or iteratively revising individual answers. We ask a prior question: \emph{what determines whether a collection of black-box responses is recoverable at all?} We represent responses to typed transformations of a query as a \emph{relational response field} (RRF). Edge transports encode how valid responses must change under paraphrase, scaling,...

    arxiv.org/abs/2608.04552 · PDF

  16. 16

    EuroExec: Frontier Language Models Fall Short of Expert Judgment on European Executive Decision Tasks

    Pau Arnal, Khaled Denfir, Danylo Smahliuk, Amrut Avhad, Marcus A. Castro

    cs.CL · cs.AI · cs.LG

    Frontier LLMs are increasingly put to use on open-ended complex questions, different in nature from the ones they are typically evaluated on. We dedicate more than 4,000 human expert hours to evaluate a selection of six frontier LLMs on a member of this class of problems: EuroExec, our introduced human expert-based benchmark composed of 413 open-ended long-form European executive tasks authored by 47 vetted domain experts, each question drawn...

    arxiv.org/abs/2608.04549 · PDF

  17. 17

    ODRA: Synthesizing Cognitive Behavioral Therapy Sessions with Structured Chain-Of-Thought and Dynamic Patient Resistance

    Javier Rodriguez-Juan, Hiba Arnaout, Jose Garcia-Rodriguez, David Tomás, Iryna Gurevych

    cs.CL · cs.LG

    Synthetic generation of Cognitive Behavioral Therapy (CBT) sessions is challenged by two competing demands: adhering to strict therapeutic structure while modeling the resistant, unpredictable behavior of real patients. Existing script-based methods fail to capture dynamic therapeutic interactions, while multi-agent approaches struggle to adhere to CBT's sequential structure; both suffer from sycophancy, producing overly compliant patients...

    arxiv.org/abs/2608.04524 · PDF

  18. 18

    D$^2$F-ReAG: Dynamic Decomposition and Filtering for Multi-Hop Reasoning-Augmented Generation

    Jiaoyang Li, Junhao Ruan, Shengwei Tang, Kaiyan Chang, Zhengtao Yu, Tong Xiao, Jingbo Zhu

    cs.CL · cs.AI

    Large language models (LLMs) often generate inaccurate answers due to their reliance on static internal knowledge. Retrieval-augmented generation (RAG) addresses this limitation by integrating external knowledge and excelling at single-hop queries. However, it struggles with multi-hop questions that require cross-document reasoning. Existing methods, such as graph structured RAG or question decomposition, often lack dynamic decomposition and...

    arxiv.org/abs/2608.04444 · PDF

  19. 19

    MERaLiON-GR: Speech Gender Recognition Model for English and SEA Languages

    Qiongqiong Wang, Ai Ti Aw, Nancy F. Chen, Ying Lay Chiu, Yang Ding, Yingxu He, Ridong Jiang, Zhuohan Liu, Yanfeng...

    cs.CL · cs.AI

    We present MERaLiON-GR, a speech gender recognition system that performs binary classification (female / male) on English and Southeast Asian (SEA) languages. The model finetunes MERaLiON-SpeechEncoder-2, a large conformer based transformer pre-trained on a broad speech corpus, and applies parameter efficient fine-tuning via Low-Rank Adaptation (LoRA) to adapt the encoder to the gender recognition task, and appends a multi-scale ECAPA-TDNN...

    arxiv.org/abs/2608.04433 · PDF

  20. 20

    FinReportBench: Measuring and Improving Institution-Grade Financial Report Generation

    Yinghao Tang, Tan Zhenwei, Yiyao Wang, Wanli Gu, Xiaolu Zhang, Jun Zhou, Wei Chen

    cs.CL · cs.AI

    Large language models can produce fluent financial analysis, but fluency alone does not establish whether a report is suitable for institutional delivery. We introduce FinReportBench, an expert-grounded benchmark for measuring and improving institution-grade financial report generation. Expert review reveals recurring gaps in report identity, institutional components, source discipline, and visual delivery. We derive a 35-item rubric through...

    arxiv.org/abs/2608.04374 · PDF

  21. 21

    Equitable System-Prompt Selection via Constrained Mixed-Strategy GroupDRO

    Mengyu Xu, Qiaoxin Yang, Zhihan Liu, Ruiyao Xu, Zachary Liu, Kezhen Chen, Chongyang Gao

    cs.CL · cs.AI · cs.CE · stat.ML

    Large language models are increasingly used for information seeking, yet semantically equivalent questions phrased in different ways can receive answers of considerably different quality. System prompts are widely employed to steer response behavior, but they are typically optimized for average-case quality, so some question phrasings may still receive incomplete or low-quality answers. To address this, we formulate a constrained...

    arxiv.org/abs/2608.04339 · PDF

  22. 22

    Right Reset: Chunking by Prefix Removal

    Mike Vegeto

    cs.CL · cs.LG

    Removing the left context from a causal language model reveals a useful kind of boundary: an edge where the model processes the same right-hand tokens with little change. We turn this observation into prefix-removal probing and introduce Right Reset (RR), which measures preservation of the right-hand hidden-state trajectory. A dynamic program converts RR edge scores into variable-length chunks. On flattened text formed by concatenating...

    arxiv.org/abs/2608.04330 · PDF

  23. 23

    MIDAS: Multi-LLM Iterative Data-Adaptive Summarization

    Karen Lee, Dhanashree Balaram, Seojun Shon, Umair Rasheed

    cs.CL · cs.AI · cs.MA

    Text summarization is deceptively difficult. While condensing information seems straightforward, real-world enterprise summarization of support tickets, legal documents, incident reports, and more, demands strict adherence to domain-specific guidelines, output formats, and organizational conventions. Crafting prompts that reliably satisfy these constraints is labor-intensive, requiring significant human expertise and continuous maintenance as...

    arxiv.org/abs/2608.04307 · PDF

This edition is part of The Daily Abstract — cs.CL archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.