cs.CL · 2026-07-29 · No. 68

Computation and Language, 2026-07-29.

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

01 — The papers

10 entries
  1. 01

    Pass the Baton: Trajectory-Relayed On-Policy Distillation

    Haolei Xu, Xiaowen Xu, Haiwen Hong, Zixuan Ni, Hongxing Li, Yiwen Qiu, Weiming Lu, Yongliang Shen

    cs.CL · cs.AI

    On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix failure: once the student commits to a wrong reasoning direction, all subsequent generation builds on this deviation, producing misdirected continuations that elicit unreliable supervision and waste compute. We identify a teacher-student continuation asymmetry on failed prefixes, where the teacher tends to redirect while the...

    arxiv.org/abs/2607.26057 · PDF

  2. 02

    Detecting Knowledge Inconsistencies Across Text, Tables, and Knowledge Graphs

    Fanfu Wei, Thibault Ehrhart, Raphaël Troncy

    cs.CL · cs.AI

    Wikipedia and Wikidata are widely used for information access, LLM pre-training, and retrieval-augmented generation. Their knowledge is deeply connected but scattered across text, tables, and knowledge graphs. This raises a practical question: when these modalities disagree, how can we detect and explain the conflict? We study this problem as \emph{modality-level inconsistency detection}. We first introduce a taxonomy of cross-modal knowledge...

    arxiv.org/abs/2607.25959 · PDF

  3. 03

    Evaluating Multi-Turn Multimodal Diagnostic Reasoning on Challenging Real-World Clinical Cases

    Rui Yang, Weihao Xuan, Yi Lin, Zhuhan Bao, Jonathan Chong Kai Liew, Matthew Yu Heng Wong, Nicolás Lescano, Nikita R....

    cs.CL · cs.AI

    Clinical diagnostic evaluation should not only assess whether models can provide correct diagnoses, but also reflect the realities of clinical practice, including progressive disclosure of multimodal information, dynamic updating of diagnostic hypotheses, and continuous refinement of clinical reasoning. However, existing evaluations of multimodal large language models (MLLMs) typically rely on single-turn or isolated tasks, making it...

    arxiv.org/abs/2607.25933 · PDF

  4. 04

    MyMentorLLM: A psychotherapy GenAI environment with multimodal voice/text patients, trainees and experts for deliberate practice

    Rodolfo Rizzi, Alessandro Grecucci, Massimo Stella

    cs.CL · cs.AI

    Psychotherapists need repeated training and supervision by experts; however, scalability is problematic. Here we present MyMentorLLM, a multimodal voice- and text-based simulation environment for deliberate practice, used to generate 2,100 complete Cognitive Behavioural Therapy (CBT) training sessions. Each session links a DSM-5-TR-grounded patient (with major depressive, generalised anxiety or borderline personality disorder), a...

    arxiv.org/abs/2607.25667 · PDF

  5. 05

    Construction-Driven Injection: Linguistically-Grounded Edit-Based Code-Mixing Fingerprints for Large Language Models

    Yongyi Cui, Yue Li, Tianbao Jiang, Xin Yi

    cs.CL · cs.AI

    Large language models (LLMs) are costly intellectual assets that remain exposed to unauthorized redistribution and commercial misuse. Injected fingerprints, i.e., trigger--target pairs embedded in model behavior, offer a practical, black-box-verifiable ownership signal, but existing methods decouple the two stages of the fingerprint life cycle: how a fingerprint is constructed and how it is injected. Existing fingerprinting frameworks suffer...

    arxiv.org/abs/2607.25633 · PDF

  6. 06

    A Human-in-the-Loop Corpus for LLM-Based Simplification of Scientific Summaries

    Kyuri Im, Michael Färber

    cs.CL · cs.AI · cs.HC

    Interdisciplinary research is accelerating, yet scientific papers remain difficult to understand outside their home fields. We study large language model (LLM)-based simplification of scientific texts and present a human-in-the-loop workflow that transforms expert summaries into more accessible versions for non-specialists. Using SciSummNet as the source corpus, we first generate baseline simplifications with GPT-4o-mini. In Phase 1, readers...

    arxiv.org/abs/2607.25630 · PDF

  7. 07

    IRIS: Reusable Identity Representations from Frozen LLMs for Entity Alignment

    Xinran Liu, Shengtao Li, Shouqian Shi, Ge Wang, Xin-Wei Yao

    cs.CL · cs.AI

    Entity alignment (EA) identifies entities across knowledge graphs (KGs) that refer to the same real-world object. Conventional EA methods mainly exploit explicit graph structures and textual fields, which often provide insufficient semantic understanding to recognize the same entity under heterogeneous descriptions and distinguish it from semantically similar entities. Although large language models (LLMs) offer deeper entity understanding,...

    arxiv.org/abs/2607.25579 · PDF

  8. 08

    Every Time I Hire a Linguist, Inference Costs Go Down: On Linguistic Rules as Effective Prompt Compressors

    Jianfei Ma, Zhaoxin Feng, Emmanuele Chersoni, Si Chen

    cs.CL · cs.AI

    Prompt compression shortens LLM input to reduce inference cost, yet existing methods score token importance through LM forward passes. It remains questionable whether such nuanced, costly token selection is necessary. Compression requires identifying informative content, a problem that linguistic research has long addressed through cues that can be operationalized as deterministic rules. We therefore ask: can \textbf{linguistic rules alone}...

    arxiv.org/abs/2607.25335 · PDF

  9. 09

    CAST: Game Solvers as Turn-Level Teachers for LLM Agents

    Yu Wang, Yi-Kai Zhang, Wentao Shi, Ziang Ye, Yuchun Miao, Yueqing Sun, Qi Gu, Xunliang Cai, Lan-Zhe Guo, Han-Jia Ye,...

    cs.CL · cs.AI

    Training large language models (LLMs) to act in long-horizon games is a promising step toward generalist decision-making, yet reinforcement learning with verifiable rewards (RLVR) relies on sparse final rewards that reveal little about which decisions determine success. Denser process signals could supply this missing turn-level credit, but existing sources are hard to keep both cheap and accurate. We observe that changes in a game solver's...

    arxiv.org/abs/2607.25308 · PDF

  10. 10

    Where Steering Signals Come From: Activation Source Selection in Activation Steering

    Jiaran Ye, Lingxu Ran, Zijun Yao, Chenpeng Wang, Yong Jiang, Lei Hou, Juanzi Li, Liangming Pan

    cs.CL · cs.AI · cs.LG

    Activation steering controls language models by adding vectors or features to hidden states at inference time, but the upstream source of these steering signals is often treated as a secondary detail. We study this source choice as activation source selection: the combination of source context and activation readout policy used to collect the hidden states from which a steering signal is built. Holding the downstream intervention fixed, we...

    arxiv.org/abs/2607.25270 · 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.