cs.CL · 2026-11-03 · No. 162

Computation and Language, 2026-11-03.

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

    Tracing the Scientific Discovery Capacity of LLMs to Their Training Data

    Arda Uzunoglu, Mahler Revsine, Michael Schatz, Benjamin Van Durme, Mehrdad Farajtabar, Daniel Khashabi, Eric Nalisnick

    cs.CL · cs.AI

    Large language models (LLMs) show growing promise for scientific discovery, yet basic questions remain about what enables reliable discovery. How can we trace the scientific discovery capacity of LLMs to their training data? We study this question as one of missing probability mass and develop a theoretical account in which monofacts-the facts observed once during training-set the discovery capacity. In materials science and regulatory...

    arxiv.org/abs/2611.03617 · PDF

  2. 02

    Can LLM Agents Be Forward-Deployed Engineers? Requirement Discovery and Delivery for Enterprise Text-to-SQL

    Marianne Menglin Liu, Rongguang Wang, Zheng Wang, Yewen Wang, Leonid Boytsov, Tao Sheng, Sujith Ravi, Dan Roth

    cs.CL · cs.AI

    LLM coding agents can increasingly automate model post-training, but whether they can act as forward-deployed engineer (FDE) agents, discovering customer requirements and turning them into deployment decisions, remains unclear. We study this on a real enterprise text-to-SQL engagement: a simulated customer, grounded in the real use case's requirements, starts with an ambiguous request. An agent must discover acceptance criteria, data...

    arxiv.org/abs/2611.03551 · PDF

  3. 03

    VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations

    Avni Mittal, Avinash Anand, Dikshant Kukreja, Erik Cambria, Rajiv Ratn Shah, Sandipan Dandapat

    cs.CL · cs.AI · cs.LG · cs.MA

    Organisations increasingly deploy large language model agents that must follow organisational values. We check whether activation steering, which adds vectors to a model's internal activations during generation without changing its weights, can improve compliance beyond prompting alone. Since most scenarios involve several values at once, we focus on multi-value steering. A simple way to build these vectors is to contrast responses that...

    arxiv.org/abs/2611.03518 · PDF

  4. 04

    Written Once, Read Differently How Language Models Own Answers Steer Their Next Ones

    Marcus Armstrong

    cs.CL · cs.AI

    When a language model answers a yes/no question, that answer changes how it answers a different question about the same entity. We study this \emph{self-commitment} in 22 open-weight models from four families (0.5B--72B parameters), fixing the model's first answer and measuring the shift in its second across 3,600 question pairs about 120 entities. Each answer writes a \emph{commitment direction} into the residual stream. The similarity of...

    arxiv.org/abs/2611.03198 · PDF

  5. 05

    An Experimental Study of Hybrid Retrieval, NLI and Open-source LLMs for Scientific Claim Verification

    Lakshmi Narayanan P, Narenkarthik K, Naren Kumar C

    cs.CL · cs.AI · cs.LG

    Claim verification is the task of deciding whether a source text supports a claim, contradicts it, or lacks sufficient evidence. Natural Language Inference (NLI) models can judge whether a single sentence supports or contradicts a claim, but they struggle when the evidence is spread across several sentences. Conversely, Large Language Models (LLMs) can reason across sentences, but may draw conclusions the text does not support. In this...

    arxiv.org/abs/2611.03093 · PDF

  6. 06

    Unlocking Criminal Court Decisions: Flexible Information Extraction for Empirical Legal Research

    Tomáš Knap, Ivana Kvapilíková, Jan Černý, Klára Bendová, Jaromír Šavelka, Vojtěch Pour, Jakub Drápal

    cs.CL · cs.AI

    Criminal verdicts contain descriptions of criminal behavior that can support quantitative legal research, but structuring them requires attributes and extraction rules that cannot always be fully specified in advance. We present an LLM-based pipeline that builds a structured dataset through attribute exploration, text span extraction, and value normalization, guided by annotation instructions that legal experts can refine. We evaluate it on...

    arxiv.org/abs/2611.03051 · PDF

  7. 07

    VERS: Steering Reasoning Models by Editing Their Activation Explanations

    Ihor Vitenko

    cs.CL · cs.LG

    Reasoning language models write a chain of thought before they answer, and many checks in it do not change the final answer. Natural language autoencoders (NLAs) explain a model's activations in text and rebuild activations from text, and editing an explanation can steer a model at a single token. We study whether such edits can also steer how a model reasons. We build VERS (Verbalize, Edit, Reconstruct, Steer): it turns explanation edits at...

    arxiv.org/abs/2611.03040 · PDF

  8. 08

    PlagiaCorpus: A Bilingual, Adversarial, and Cross-Lingual Dataset for Forensic Plagiarism Detection

    Adel Ammar, Serry Sibaee, Wadii Boulila

    cs.CL · cs.AI · cs.CY · cs.DL · cs.IR

    PlagiaCorpus is a bilingual Arabic-English dataset for benchmarking plagiarism and text-reuse detection, including reuse produced by large language models (LLMs) and reuse across languages. It contains 198,643 text records: a source corpus of 34,885 passages drawn from English and Arabic Wikipedia and two openly licensed book collections, and 163,758 queries (147,381 for training and 16,377 for testing). The queries cover five reuse classes...

    arxiv.org/abs/2611.02940 · PDF

  9. 09

    Maswao: An automatically collected, prepared, and benchmarked dataset for the South African Sign language

    Thapelo Sindane, Unarine Netshifhefhe, Vukosi Marivate, Abiodun Modupe, Herkulaas Combrink, Ketro Sithole

    cs.CL · cs.AI

    Recent advancements in sign language processing can be attributed to developments in both large-scale dataset collection and algorithmic complexity in areas such as pose estimation techniques, signal processing through Transformer models, and model optimization. However, only a handful of sign languages, such as the American, British, and German, from over 200 globally recognized sign languages, reap the benefits of such advancements, while...

    arxiv.org/abs/2611.02903 · PDF

  10. 10

    Tug of Watts: A study on the Energy - Accuracy Trade-off of Large Language Models on Medical Question Answering Tasks

    Viju Sudhi, Giulio Billi, Andreea Scrob, Regina Stodden, Philipp Cimiano

    cs.CL · cs.AI

    Large Language Models (LLMs) are increasingly used in medical question answering (QA) systems. While recent works aim at incrementally improving the downstream task accuracy, an empirical estimation of the models' energy consumption and the associated carbon footprints is often overlooked. In this study, we demonstrate how practitioners can estimate energy consumption and arrive at reliable emission estimates for medical QA tasks available in...

    arxiv.org/abs/2611.02889 · 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.