cs.CL · 2026-10-10 · No. 139

Computation and Language, 2026-10-10.

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

01 — The papers

8 entries
  1. 01

    Predicting Alignment Generalization with Value Representations

    Andy Liu, Mehar Bhatia, Karolina Stanczak, Mona Diab, Vered Shwartz, Daniel Fried

    cs.CL · cs.AI · cs.LG

    LLM developers post-train their models to exhibit prosocial values and behavioral traits, which are enumerated in an alignment target. However, while recent post-training developments have yielded models that score highly on alignment evaluations, training models on sets of narrow behaviors still influences their behavior across unseen contexts and environments in unexpected ways. In this paper, we establish the task of alignment...

    arxiv.org/abs/2610.12410 · PDF

  2. 02

    VFold: Symmetry-Aware Cross-Layer Value Cache Compression

    Neha Verma, Sungwon Kim, Kenton Murray, Kevin Duh

    cs.CL · cs.LG

    While caching key-value (KV) states accelerates Large Language Model (LLM) decoding, this cache can dominate memory usage at long context lengths. One solution is to compress this memory by exploiting inter-layer cache similarities. However, most existing techniques necessitate architectural changes to LLMs and incur substantial overhead. In this work, we propose a symmetry-aware value cache merging strategy that reduces cache memory while...

    arxiv.org/abs/2610.12338 · PDF

  3. 03

    SciTBERT: A family of chronologically consistent language models for scientific and technological language processing

    Thomas Gebhart, Russell J. Funk

    cs.CL · cs.LG

    Pre-trained transformer models are increasingly being used to study scientific and technological progress. Encoders tuned to paper or patent text outperform general-purpose models on downstream classification, regression, and proximity tasks within science and technology. However, the applicability of these models for studying time-dependent or archival properties of science, technology, and their interface is limited due to lookahead and...

    arxiv.org/abs/2610.12207 · PDF

  4. 04

    Rehearse Everything, Remember Nothing: Attic-KV Rehearses What Will Be Read

    Zhiyun Shi

    cs.CL · cs.LG

    Many key-value (KV) caches are compressed before anyone knows what will be asked of them: a document cached for retrieval, a prompt prefix shared across requests, the memory of a long conversation. The prevailing approach scores KV entries by rehearsal: the model rereads the context and keeps the entries it attends to, assuming that the more completely a cache rehearses its context, the better it remembers it. We show that under tight budgets...

    arxiv.org/abs/2610.12133 · PDF

  5. 05

    All Verdicts are Not Equal: Rethinking LLM Judge Reliability

    Vineet Kumar, Darshita Rathore, Anindya Moitra

    cs.CL · cs.AI

    LLM-as-a-Judge is the standard paradigm for NLP evaluation, yet its systemic reliability remains poorly understood despite being widely treated as a deterministic ground truth. We present a comprehensive reliability audit, stresstesting six frontier models across four benchmarks, five prompt formats, two presentation orders, three sampling temperatures, and ten repetitions per condition. Our empirical analysis reveals severe vulnerabilities:...

    arxiv.org/abs/2610.12083 · PDF

  6. 06

    Examining Social Attribution in LLM Reasoning: A Theory-Guided Probing Methodology

    Zhaoxin Yu, Qingchao Kong, Dajun Zeng, Wenji Mao

    cs.CL · cs.AI · cs.LG

    Large language models (LLMs) are increasingly deployed in sociotechnical systems where social attribution, the reasoning process attributing external events to the causes and reasons of agents' social behaviors, plays a critical role. These processes involve judgments of social cause, responsibility, and blame/credit to agents. Although attributional models are well-studied in social psychology and cognition through Attribution Theory, social...

    arxiv.org/abs/2610.12022 · PDF

  7. 07

    Forms of LLM-Integrated Applications from LLM-Chats to Autonomous AI Agent System

    Irene Weber

    cs.CL · cs.AI · cs.SE

    Large language models (LLMs) are increasingly embedded as components in software systems, marketed under labels such as chatbot, copilot, retrieval-augmented generation, workflow, coding agent and AI agent. Whether these labels denote genuine architectural forms or serve as branding has not been assessed systematically. In the sources surveyed, labels do carry architectural content, most clearly in vendor usage: copilot denotes a...

    arxiv.org/abs/2610.11899 · PDF

  8. 08

    Same Outcome, Different Evidence: Intent Recovery in LLM Safety Evaluation

    Haitong Jiang, Chunlin Liu, Sihan Tang, Chan Wu, Xiaoqing Su, Yuhong Feng

    cs.CL · cs.AI

    Safety evaluations of large language models commonly summarize harmful-output behavior with attack success rate (ASR). Yet the same non-harmful outcome can arise for very different reasons. A model may recover a harmful task and refuse it, fail to recover the task, or respond to something else entirely. Distinguishing these cases becomes especially important under intent-obscuring prompts, where a low ASR does not reveal whether the evaluated...

    arxiv.org/abs/2610.11766 · 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.