cs.CL · 2026-10-23 · No. 151

Computation and Language, 2026-10-23.

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

01 — The papers

22 entries
  1. 01

    Low-Bit N-Gram Embeddings for Language Models

    Stephen Chung, Wenyu Du

    cs.CL · cs.AI

    Many recent open-weight language models use n-gram embedding tables to improve performance. These tables store learned representations of short token sequences, such as pairs or triples of consecutive tokens, which the model retrieves during processing. However, they can require substantial memory: DeepSeek-V4.1-Flash, for example, allocates 196 billion parameters to its Engram n-gram memory. We therefore investigate whether this memory cost...

    arxiv.org/abs/2610.31646 · PDF

  2. 02

    How Ground Truth Shapes Retrieval Scores: Evidence-Aware Evaluation of a Retrieval-Augmented Generation System for Breast Lesions

    Miguel Mochizuki Silva, Samuel Ismael, Mateus C. Dias, Thaís Gaudêncio, Yuri Malheiros, Barufaldi

    cs.CL · cs.AI

    Purpose: To describe an evidence-aware retrieval evaluation protocol for a clinical retrieval-augmented generation (RAG) system over breast lesion management literature, addressing the dependence of retrieval scores on annotation coverage. Methods: An 18-document corpus and an 18-query development benchmark, with one Yes/No or WH-question per document, were used. Retrieval combined dense embeddings, BM25, and cross-encoder reranking. Chunk...

    arxiv.org/abs/2610.31606 · PDF

  3. 03

    Would an LLM Match into a Residency? Uncertainty, Reasoning and Image Use of Open Language Models on the Spanish MIR Examination

    Miguel Fernandez-de-Retana, Unai Zulaika, Ruben Sanchez-Corcuera, Aritz Bilbao-Jayo, Aitor Almeida

    cs.CL · cs.AI · cs.CY · cs.LG

    Accuracy on medical examinations is common evidence offered for the clinical use of language models, yet a health system that intends to run an open model on its own hardware needs to know considerably more: whether the model recognizes when it is likely to be wrong, whether medical fine-tuning improves this ability, whether its correct answers rest on sound clinical reasoning, or whether it actually uses the images it is given. We examine...

    arxiv.org/abs/2610.31586 · PDF

  4. 04

    The Geometry of Metaphor: Probing and Steering Metaphor Representations in Large Language Models

    Shihui Li, Mette Andersen, Martha Lewis

    cs.CL · cs.AI

    Metaphor is a crucial aspect of human cognition. As such, we would like to understand how it is represented in large language models (LLMs) and how it might be controlled. We use mechanistic interpretability techniques to identify causal subspaces in a range of LLMs covering different architectures and sizes. We show that we are able to steer output on classification tasks across different parts of speech. We are also able to steer output on...

    arxiv.org/abs/2610.31520 · PDF

  5. 05

    How Depth Growth and Looping Shape In-Context Learning in Language Models

    Ferdinand Kapl, Kaitlin Maile, Stefan Bauer, Johannes von Oswald

    cs.CL · cs.AI · cs.LG

    Looped and depth-grown language models reuse computations across depth and have attracted interest for their reasoning capabilities. Building on prior suggestive evidence of improved in-context-learning (ICL), we investigate how such reuse favors ICL, including how it shapes the acquisition, retention, and mechanisms of this ability. Our analysis spans pretrained language models up to 3B parameters and controlled tasks. The experiments...

    arxiv.org/abs/2610.31421 · PDF

  6. 06

    Same Entity, Different Script: Diagnosing Cross-Script Entity Matching Failures in Language Models

    Athira Gopal, Patil Prathamesh Pravin, Ankush Kadu, Arpit Jain, Ashwanth Krishnan

    cs.CL · cs.AI

    Multilingual language models can give different identity judgments when the same people's names are written in different scripts. We study this effect using 601 audited identity pairs with attested Devanagari and Latin forms, holding the question and relation fixed. Qwen2.5-7B's same-person accuracy falls from 35.1% in Latin to 8.5% in Devanagari under Marathi prompts, while different-person accuracy remains above 98%. Evaluation across five...

    arxiv.org/abs/2610.31407 · PDF

  7. 07

    HERD: Harness-Evolving Recursive Distillation for Continual Learning

    Xun Sun, Thanh Thi Nguyen, Thuy-Trang Vu

    cs.CL · cs.AI · cs.LG

    Continual learning requires a model to keep acquiring new skills and knowledge without compromising its existing capabilities. Parametric methods suffer from catastrophic forgetting and loss of plasticity. Non-parametric methods, such as prompts, skills, and harnesses, leverage in-context learning but are limited by context length and lossy plain-text compression. To effectively combine their complementary strengths, we propose HERD...

    arxiv.org/abs/2610.31401 · PDF

  8. 08

    SurveyGraph: Generating Academic Surveys via Methods-Entity Knowledge Graphs

    Yifan Yuan, Yancheng Zhu, Zuoyu Zhang, Xiaogang Peng

    cs.CL · cs.AI · cs.IR

    Most automated survey generation systems retrieve papers as flat documents, limiting their ability to capture research structure. We present SurveyGraph, a framework built around a Methods-Entity Knowledge Graph (MEKG) encoding method-level follow-on, competition, and shared-resource relations. The graph supports core-paper discovery, retrieval, hybrid outlining, and evidence-gated narrative synthesis. We evaluate retrieval, semantic edges,...

    arxiv.org/abs/2610.31317 · PDF

  9. 09

    When LLMs Forget What They Believe: Characterizing State Integrity Collapse in Long-Horizon Dialogues

    Yancheng Zhu, Yifan Yuan

    cs.CL · cs.AI

    Large language models can repeat an instruction while failing to keep obeying it. We study this recall-compliance dissociation in long-horizon dialogue and define State Integrity Collapse: a model preserves a stated constraint in memory but stops applying it to later turns. We introduce StateTrack-Bench, a controlled suite of conversations covering numerical, format, role/persona, factual-memory, and logical-rule constraints. Across seven...

    arxiv.org/abs/2610.31314 · PDF

  10. 10

    CoWriter: Guiding Methodological and Synergetic Human-LLM Co-Authoring

    Shmuel Amar, Benjamin Lieber, Ori Shapira, Debanjana Kar, Sebastian Maximilian Dennerlein, Yufang Hou, Ido Dagan

    cs.CL · cs.AI · cs.HC

    While LLMs have transformed text generation, standard chat-and-canvas interfaces often reduce human authoring to quite passive supervision, causing review fatigue, over-reliance on AI, and the sidelining of the human role in long-form, source-grounded writing. Prior systems typically address isolated writing phases, leaving comprehensive end-to-end human-LLM collaboration largely unexplored. In this paper, we introduce CoWriter, a proactive...

    arxiv.org/abs/2610.31277 · PDF

  11. 11

    Memory-Aware Reward Modeling: Learning Human Preferences for LLM Memory Use

    Yijiang River Dong, Zexue He, Cheng Yu, Tiancheng Hu, Michael J Ryan, Jan Batzner, Zheng Hui, Shaolong Wu, Orestis...

    cs.CL · cs.AI

    Memory enables language model assistants to maintain continuity across sessions and generate personalized responses, but knowing more about a user does not necessarily produce responses they prefer. Existing evaluations assess memory capabilities without directly establishing whether memory use benefits the user. To study this gap, we collect approximately 1,000 pairwise judgments from 87 participants using their own ChatGPT queries and...

    arxiv.org/abs/2610.31272 · PDF

  12. 12

    When More Metadata Misleads: Probing Field-Level Grounding in LLM-Generated Museum Descriptions

    Clemencia Siro, Laura Hollink

    cs.CL · cs.AI · cs.DL · cs.LG

    Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields should improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across geographic regions. We study how metadata composition shapes the generation of artefact descriptions through a two-phase framework. Phase 1 analyses natural metadata...

    arxiv.org/abs/2610.31193 · PDF

  13. 13

    Prompting Strategy Effects Depend on the Judge: A Multilingual Study of LLM-as-a-Judge Robustness

    Asal Mehrabi, Mohammad Aliannejadi, Clemencia Siro

    cs.CL · cs.AI · cs.LG

    LLM-as-a-Judge evaluation is increasingly applied to multilingual tasks, yet prompting strategies such as Chain-of-Thought and Role Conditioning are recommended based solely on English benchmarks. Whether their effectiveness transfers across judge models and languages remains unknown: prompt-sensitivity studies are largely English-centric, while existing multilingual judge studies fix the prompt template. In this work, we address this gap by...

    arxiv.org/abs/2610.31172 · PDF

  14. 14

    Prompt Sensitivity in Tool-Augmented Model Comparisons

    Mateusz Modrzejewski, Tomasz Radzikowski

    cs.CL · cs.AI · cs.LG

    Language models are compared on question-answering benchmarks that pose each question in one wording, and a reworded question can change which of two models scores higher. We ask whether giving the models tools makes such comparisons more stable. An accuracy gain cannot answer this, because tools change several steps between the question and the scored answer at once. We propose a matched protocol that separates those steps: the answer...

    arxiv.org/abs/2610.31148 · PDF

  15. 15

    Mixing Matters: Evidence Position Bias Across Sequence Mixers in Long-Context Question Answering

    Aman Behera, Namit Solanki, Mehul Anand

    cs.CL · cs.AI · cs.LG

    Long-context models often use answer-bearing evidence less effectively when it appears away from a prompt boundary. We compare this position bias across Transformer, state-space, and hybrid Mamba architectures by moving one gold document through a fixed set of nine distractors. The evaluated Transformers show a beginning advantage that the evaluated state-space models lack, while both families favor evidence near the end. A matched...

    arxiv.org/abs/2610.31135 · PDF

  16. 16

    Delta N-gram: Forward-Only Continual Writing into a Language Model's Native N-gram Memory

    Hao Wu

    cs.CL · cs.LG

    Our previous analysis found that a model's native hashed N-gram table is where continual updates disturb the base least, but its writes needed backpropagation and changed the pretrained table. We present Delta N-gram, a forward-only write path into that table. The pretrained table T^0 is never modified. Evidence is written into a sparse patch Δon the rows it addresses: a 4.1M-parameter writer emits a target and a gate per (position, hash...

    arxiv.org/abs/2610.31125 · PDF

  17. 17

    A Grammar of French Civil Cassation: A Ground-Level Dataset and Tool

    Avrile Floro, Tamara Dhorasoo, Nils Holzenberger, Thomas Le Goff, Tiphaine Viard, Maria Boritchev

    cs.CL · cs.AI · cs.CY

    This paper presents a dataset of 121,536 ground-reasons pairs across 86,464 rulings of the five civil chambers of the French Court of Cassation (2016-2025). An appeal on points of law cannot itself be appealed and often decides a case for good. Understanding how the Court rules is thus crucial and can help lawyers who draft those appeals. Inside a ruling, every ground of appeal (a distinct legal complaint against the appealed decision) has...

    arxiv.org/abs/2610.31116 · PDF

  18. 18

    What Pairwise Evaluation Misses: Reward Scores across Prompts

    Mateusz Modrzejewski, Tomasz Radzikowski

    cs.CL · cs.AI

    Reward models score responses for preference optimization, filtering and ranking, and pooling scores across prompts needs them to be comparable. Yet pairwise evaluation compares two responses to the same prompt, so any score component shared by a prompt's responses cancels and goes untested. We study this component by comparing between-prompt variation in model scores with the same variation in human ratings of the same responses. We propose...

    arxiv.org/abs/2610.31090 · PDF

  19. 19

    Harm Is Behavioural, Moderation Is Lexical: How Language and Culture Decide Whom Automated Safety Systems Protect and Whom They Silence

    Gabriel Lars Sabadin, Dr Nicola Harding, Sarah Kingston

    cs.CL · cs.AI

    Content warning: this paper quotes slurs and hateful statements from published test suites; the racial slur among them is masked. Automated systems decide at scale which messages are harmful, and they decide unevenly. Classifiers label African American English as offensive up to twice as often as other English, the accounts of prominent drag queens can score as more toxic than those of white nationalists, and safety behaviour that holds in...

    arxiv.org/abs/2610.31047 · PDF

  20. 20

    Mostly Reading: What Open-Book Scores Measure in Document-Level LoRA Adapters

    Martino M. L. Pulici, Cuong Xuan Chu, Evgeny Kharlamov, Volker Tresp

    cs.CL · cs.AI · cs.LG

    Single-document LoRA adapters have been proposed as units of knowledge to be merged and retrieved; yet open-book scoring, with the document in the prompt, can credit a document's own adapter (its specialist) for reading that document as readily as for holding it. Training one adapter for each of 150 Wikipedia articles on Llama-3.1-8B-Instruct, we find that open-book scores mostly measure a reading behaviour every adapter shares: a random...

    arxiv.org/abs/2610.31017 · PDF

  21. 21

    Joint or Per-Option? Compact Encoders for Closed-Set Decisions from 17 Million to 1 Billion Parameters

    Tardelli Ronan Coelho Stekel

    cs.CL · cs.LG

    Many language-model pipelines end in a closed-set decision, such as selecting a tool. Prompting a large language model for it is slow and order-sensitive. We ask how small an encoder for such decisions can be and whether it should read the options jointly or separately. In a three-seed study at 32 million parameters, per-option scoring beats joint scoring on a held-out sample of a 37-benchmark decision suite by 5.8 points of chance-corrected...

    arxiv.org/abs/2610.30970 · PDF

  22. 22

    RetiJev: Fast and Source-Grounded Ophthalmic Decision Models

    Bo Liu, Lianyu Wang, Meng Wang, Ching-Yu Cheng, Huazhu Fu

    cs.CL · cs.AI

    Ophthalmic clinical decision support requires fast, accurate, and privacy-preserving inference, yet autoregressive large language models (LLMs) incur substantial latency and computational overhead for decisions among predefined actions. We therefore investigate a decision-oriented alternative to generative reasoning. Specifically, we first introduce RetiJev-Data, a source-grounded ophthalmology corpus comprising 78,158 clinical-decision and...

    arxiv.org/abs/2610.30947 · 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.