cs.CL · 2026-07-22 · No. 61

Computation and Language, 2026-07-22.

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

01 — The papers

19 entries
  1. 01

    Copy Less, Ground More: Overcoming Repetitive Copying in Long-Context Reasoning via Evidence-Aware Reinforcement Learning

    Lizhe Fang, Weizhou Shen, Tianyi Tang, Yisen Wang

    cs.CL · cs.AI

    Large language models that generate step-by-step reasoning traces have achieved strong performance on complex tasks, and extending them to long-context settings has emerged as an important frontier. However, we identify a critical failure mode in this regime: \emph{repetitive copying}, where models extensively copy text from the input into their reasoning traces rather than productively solving the problem. We show that this behavior is...

    arxiv.org/abs/2607.19345 · PDF

  2. 02

    Prompt Design at Scale: How Format, Instruction Count, and Context Length Shape Instruction Adherence and Hallucination in Large Language Models

    Netanel Eliav

    cs.CL · cs.AI

    Practitioners make three prompt-design decisions with almost no controlled evidence behind them: how to format instructions and context (markdown, plain text, prose, or tabular), how many simultaneous instructions a system prompt can carry before compliance degrades, and how much context a model can hold before recall and honesty degrade. We report two controlled experiments crossing all three factors on one held, contamination-free synthetic...

    arxiv.org/abs/2607.19257 · PDF

  3. 03

    Inference-Time Steering for Cross-Lingual Factual Consistency in LLMs

    Alexander Manev

    cs.CL · cs.AI

    Although Large Language Models (LLMs) demonstrate remarkable multilingual fluency, their internal knowledge representations remain disproportionately biased toward high-resource languages. This leads to cross-lingual factual inconsistency, where they shift their empirical answer distributions based solely on the prompt language. We investigate whether these biases can be mitigated at inference time, forcing an English-prompted model to answer...

    arxiv.org/abs/2607.19243 · PDF

  4. 04

    The Price of Reasoning: Cost-Quality Tradeoffs in Reinforcement Learning for Neural Machine Translation

    Michael Jungo, Aixiu An

    cs.CL · cs.AI · cs.LG

    Reinforcement learning with verifiable rewards (RLVR) has been established as a viable paradigm for the post-training of Large Language Models (LLMs), including downstream tasks, such as Neural Machine Translation (NMT). With the latest research indicating that RLVR could be the preferred training method for translating legal documents due to the induced reasoning capabilities, it raises the question whether it is really attributed to the...

    arxiv.org/abs/2607.19226 · PDF

  5. 05

    Beyond Score Prediction: LLM-Based Essay Scoring and Feedback Generation via Reinforcement Learning with Rubric Rewards

    Xuefeng Jin, Jiashuo Zhang, Teng Cao, Bin Yang

    cs.CL · cs.AI

    Large language models (LLMs) have been widely applied to automated essay scoring (AES) and automated feedback generation (AFG). However, existing studies rely primarily on prompt engineering or supervised fine-tuning, while systematic research on reinforcement learning (RL) post-training and automated evaluation of feedback quality remains limited. We propose RLAES, a unified LLM framework that jointly optimizes essay scoring and feedback...

    arxiv.org/abs/2607.19219 · PDF

  6. 06

    MIRA-Ev:A Benchmark for Granular Evidence Detection and Relational Reasoning in Clinical Exams

    Iker De la Iglesia, Johanna Ramirez-Romero, Jose Maria Villa-Gonzalez, Irune Urroz García, Ander Barrena, Aitziber Atutxa

    cs.CL · cs.AI

    Clinical NLP evaluation remains dominated by multiple-choice question answering (MCQA), which scores only final-answer accuracy and cannot detect when a model reaches the correct diagnosis while grounding it in irrelevant, absent, or contradictory evidence. We introduce MIRA-Ev, a clinical argument mining benchmark built on Spanish Médico Interno Residente (MIR) licensing-exam cases, re-annotated by expert clinicians with span-level premises,...

    arxiv.org/abs/2607.19201 · PDF

  7. 07

    Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning

    Aixiu An, Michael Jungo, Eloi Eynard, Mark Drenhaus, Andreas Fischer, Jean Hennebert, Sébastien Rumley

    cs.CL · cs.AI · cs.LG

    Neural machine translation (NMT) in the legal domain is a linguistically and conceptually demanding task, primarily due to the complexity of legal language and the high level of precision it requires. The recent emergence of reasoning-capable language models opens new possibilities for tackling such challenges. They add to a set of other previously proposed techniques to enhance the translation quality, which includes supervised fine-tuning...

    arxiv.org/abs/2607.19181 · PDF

  8. 08

    Translation as Augmentation: Effect of Translated Data on Assessment of Difficulty

    Yiheng Wu, Jue Hou, Roman Yangarber

    cs.CL · cs.LG

    Reliable Text Difficulty Assessment is a prerequisite for valid text simplification workflows and personalized learning applications. However, the development of robust assessment models is severely hindered by a critical bottleneck: the scarcity of expert-annotated corpora containing fine-grained difficulty levels (e.g., CEFR), particularly for lower-resource languages. This paper addresses this data scarcity problem in the context of a...

    arxiv.org/abs/2607.19101 · PDF

  9. 09

    DAIS: Dependency-Aware Intermediate QA Supervision for Complex Reasoning

    Yu Wang, Ming Fan, Xicheng Zhang, Zhiyong Li, Zhihu Wang, Caiyue Xu, Dahai Hu, Ting Liu

    cs.CL · cs.AI

    Chain-of-thought (CoT) supervision exposes intermediate rationales, but flat rationale targets usually optimize a single reasoning sequence and provide limited supervision on how local conclusions should support later decisions. We introduce Dependency-Aware Intermediate QA Supervision (DAIS), a training-time framework that converts filtered teacher rationales into stage-level QA records. Each intermediate record predicts a local answer...

    arxiv.org/abs/2607.19088 · PDF

  10. 10

    Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges

    Tuo Liang, Zhe Hu, Disheng Liu, Jing Li, Yu Yin

    cs.CL · cs.AI · cs.MM

    Multimodal humor in memes, cartoons, and comics remains difficult for AI systems because intended meaning depends on non-literal mechanisms, shared cultural knowledge, and communicative intent rather than literal scene description. This survey focuses on visual humor understanding in single-image and multi-panel artifacts, while treating humor generation as an emerging downstream frontier. We position the literature against prior humor,...

    arxiv.org/abs/2607.19011 · PDF

  11. 11

    MedDDC-Eval: Diagnosis-Decoupled Evaluation of Multi-Turn Medical Consultation Agents

    Guofeng Zhang, Yizeng Quan, Huaiyi Fang, Jianwei Lv, Jinyao Liu, Xunxu Duan, Lening An, Yu Ouyang, Junfeng Wang

    cs.CL · cs.AI

    Multi-turn medical consultation agents must decide what to ask, adapt to patient responses, and determine when the collected evidence is sufficient. However, coupled evaluation conflates the quality of the policy-elicited history with policy-specific terminal diagnosis generation: strong generation can compensate for a thin history, while weaker generation can obscure a rich one. We introduce MedDDC-Eval, a diagnosis-decoupled testbed that...

    arxiv.org/abs/2607.18999 · PDF

  12. 12

    AutoJourn: Multi-Perspective Summarisation, Bias Detection and Bias Neutralisation for LLM-Generated News in Automated Journalism

    Himel Ghosh, Ahmed Mosharafa, Georg Groh

    cs.CL · cs.AI · cs.HC

    We present AutoJourn, a demonstration system for multi-perspective news generation and bias-aware evaluation using large language models (LLMs). The system tackles three core challenges in responsible automated journalism: extracting diverse perspectives from unstructured social media discussions, generating summaries that preserve viewpoint diversity, and detecting or mitigating bias in AI-generated news. The pipeline integrates advanced...

    arxiv.org/abs/2607.18983 · PDF

  13. 13

    Verifiable Self-Evolution for Open-Ended Dialogue Skills via Future-Feedback Prediction

    ChaoJin Zhao, Xuan Jiang

    cs.CL · cs.AI · cs.LG

    Textual skills provide a lightweight way to improve frozen language-model agents, but their self-evolution normally requires a stable validation signal. Such signals are natural in mathematics or code, where an answer can be checked after it changes, yet are problematic in open-ended dialogue: changing the assistant response also changes the user's next reaction, so a logged reaction cannot directly evaluate a counterfactual response. We...

    arxiv.org/abs/2607.18973 · PDF

  14. 14

    AILQA: Evaluating AI-Driven Legal Question Answering Systems for the Indian Legal System

    Shubham Kumar Nigam, Shubham Kumar Mishra, Noel Shallum, Kripabandhu Ghosh, Arnab Bhattacharya

    cs.CL · cs.AI · cs.IR

    This comprehensive study introduces an advanced Artificial Intelligence for Indian Legal Question Answering (AILQA) system tailored to the Indian legal context. AILQA leverages a variety of embedding and generative models, including recent Large Language Models (LLMs), to address the unique challenges posed by the intricate and diverse nature of Indian legal texts and to enhance the accuracy and reliability of responses to legal questions. We...

    arxiv.org/abs/2607.18825 · PDF

  15. 15

    Find Before You Fine-Tune: A Diagnostic Study of Small LLMs for Cybersecurity QA

    Shaswata Mitra, Subash Neupane, Trisha Chakraborty, Himanshu Tripathi, Sudip Mittal, Aritran Piplai, Shahram Rahimi

    cs.CL · cs.AI · cs.CR

    Large Language Models (LLMs) are increasingly fine-tuned for critical-domain Question-Answering (QA), yet choosing which small model to adapt, before paying the cost of adaptation, remains difficult. Fine-tuning can improve domain alignment, but it may also erode prior knowledge, weaken instruction-following, or increase hallucination, especially when labeled data are scarce or rapidly evolving as in cybersecurity. We present FiT (Find before...

    arxiv.org/abs/2607.18725 · PDF

  16. 16

    LatentMT: Machine Translation with Latent Reasoning

    Wei-Rui Chen, Samar M. Magdy, Chiyu Zhang, Wenhui Zhu, Zhipeng Wang, Muhammad Abdul-Mageed

    cs.CL · cs.AI · cs.LG

    Latent-reasoning looped language models (LoopLMs) offer a different scaling path for machine translation (MT): instead of increasing parameter count or emitting explicit chain-of-thought tokens, they spend additional recurrent computation inside hidden states. We introduce LatentMT, the first systematic study of latent-reasoning LoopLMs for machine translation. LatentMT adapts a small 2.6B-parameter backbone model with lightweight training....

    arxiv.org/abs/2607.18618 · PDF

  17. 17

    Stochastic Meta-Unlearning: Bridging Language Backbone and Multimodal Unlearning

    Zijie Liu, Jinhao Duan, Gaowen Liu, Sijia Liu, Tianlong Chen

    cs.CL · cs.LG

    Machine unlearning for vision-language models (VLMs) remains underexplored. Unlike language models, VLMs combine a language backbone with visual components, which makes unlearning more complex. There is a surprising phenomenon when moving from single-modality unlearning to VLM unlearning: a target forgotten by the standalone language backbone can still be recovered when image information is given to the full VLM. This shows that text-only...

    arxiv.org/abs/2607.18615 · PDF

  18. 18

    For What Reason? Interpreting Models' Encoding of Causation and Antithesis

    Abhidip Bhattacharyya, Shira Wein

    cs.CL · cs.AI

    Discourse relations provide document structure, critical to language understanding and enabling language model performance and ethicality. In this work, we investigate how instruction-tuned Transformer models (LLaMA and Mistral) encode discourse relations in English, with a particular focus on the contrasting relations of causation and antithesis. Framing the task as a next-token prediction task and applying a suite of interpretability...

    arxiv.org/abs/2607.18570 · PDF

  19. 19

    The Story Shapes the Agent: Narrative Priors in LLM Behavior

    Yixuan Wang, James Lester, Shashank Srivastava

    cs.CL · cs.AI

    Persona prompting is widely used to steer LLM agent behavior, yet the narrative framing of a task can matter more than the assigned persona. We isolate this effect through structural isomorphism, constructing three text-based investigation games that share the same action space, stage progression, and resource constraints while varying only task narrative: disease investigation, IT troubleshooting, and murder mystery. Across 1,890 sessions...

    arxiv.org/abs/2607.18566 · 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.