cs.CL · 2026-07-13 · No. 52

Computation and Language, 2026-07-13.

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

01 — The papers

14 entries
  1. 01

    Task-Specific Multimodal Question Answering Agents via Confidence Calibration and Incremental Reasoning for QANTA 2026

    Nirjhar Das, Md. Al-Mamun Provath

    cs.CL · cs.AI

    We present our submission to the QANTA 2026 shared challenge at the ICML 2026 Workshop on Efficient Multimodal Question Answering (EMM-QA). Quanta evaluates multimodal quizbowl systems that answer pyramid-style questions from incrementally revealed text and accompanying images while operating under realistic efficiency constraints. The challenge consists of two distinct tasks: Tossup questions, which require deciding when to answer under...

    arxiv.org/abs/2607.09623 · PDF

  2. 02

    Conceptual Networks for Cross-Linguistic Idiomatic Expressions:A Feature-Based Graph Approach

    Kiran Pala, Punam Silu, Lixun Yu

    cs.CL · cs.AI · cs.ET

    We present an interpretable network-based framework for representing idiomatic and figurative meaning across eight typologically diverse languages, totaling 160 conventional expressions, the large majority of which are idiomatic. Each expression is annotated with binary conceptual features (containment, concealment, emotional, social, etc.) derived from cognitive-linguistic theory, and pairwise Jaccard similarities define a weighted graph....

    arxiv.org/abs/2607.09576 · PDF

  3. 03

    Test-Time Scaling for Small VLMs on Multilingual Visual MCQ

    Spiros Baxevanakis, Peng-Jian Yang

    cs.CL · cs.AI · cs.LG

    Test-time scaling (TTS) reliably improves reasoning in large language models, but whether it transfers to small open vision-language models remains unclear. We examine this on EXAMS-V, a multilingual visual multiple-choice benchmark, comparing self-consistency, describe-then-reason with PRM-guided beam search, and two post-hoc selectors across Qwen2.5-VL-7B-Instruct and Qwen3.5-4B. What matters is the conditions under which TTS runs, not the...

    arxiv.org/abs/2607.09438 · PDF

  4. 04

    A Sovereign, Open-Source Foundation Model for German and English

    The Soofi-Team, :, Benedikt Droste, David Fitzek, Ruben Härle, Lukas Helff, Maximilian Idahl, Alex Jude, Abbas Goher...

    cs.CL · cs.AI · cs.LG

    We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English. Its hybrid design activates only 3B of 30B parameters per token and keeps the inference cache near-constant as context grows, giving it a decisive throughput advantage over dense models for long-context, high-concurrency deployment. Pretrained on roughly 27 trillion tokens with deliberately...

    arxiv.org/abs/2607.09424 · PDF

  5. 05

    Self-Guided Test-Time Training for Long-Context LLMs

    Xinyu Zhu, Zhe Xu, Xiaohan Wei, Yunchen Pu, Fei Tian, Chonglin Sun, Kaushik Rangadurai, Hua Zhi, Frank Shyu, Sandeep...

    cs.CL · cs.AI

    Long-context processing has become increasingly important for large language models (LLMs), but simply extending the context window does not guarantee effective utilization of long inputs. As input length grows, accuracy often degrades, indicating that models still struggle to identify and use the evidence most relevant to a question. A promising way to improve long-context utilization is test-time training (TTT), which treats the test...

    arxiv.org/abs/2607.09415 · PDF

  6. 06

    Deceptive Grounding: Entity Attribution Failure in Clinical Retrieval-Augmented Generation

    Cedric Caruzzo, Donggeun Yoo, Tae Soo Kim

    cs.CL · cs.AI · cs.LG

    Retrieval-augmented generation evaluation checks whether model claims are factually grounded in retrieved documents. It does not check whether retrieved evidence is attributed to the correct entity. A clinical RAG response can pass every automated check (zero hallucinations, near-perfect faithfulness, real citations) while presenting drug Y's clinical evidence as evidence about queried drug X. We term this deceptive grounding (DG): a failure...

    arxiv.org/abs/2607.09349 · PDF

  7. 07

    WILDTRACE: Benchmarking Natural Evidence Trails in Long-Context Reasoning

    Zixin Chen, Peng Liu, Haobo Li, Rui Sheng, Jianhong Tu, Xiaodong Deng, Fei Huang, Kashun Shum, Dayiheng Liu, Huamin Qu

    cs.CL · cs.AI

    Answering complex questions over long documents frequently requires integrating evidence that the source itself disperses naturally across distant passages. In an incident report, the operating condition, design flaw, and missed safety check that jointly explain a disaster may appear dozens of sections apart; in a novel, a character's true motive may surface only through scenes far removed from the moment it becomes relevant. This...

    arxiv.org/abs/2607.09328 · PDF

  8. 08

    Letting the Data Speak: Extracting Keywords from Crowdsourced Collections with AI

    Miguel Arana-Catania, Catherine Conisbee, Matthew Kidd

    cs.CL · cs.AI · cs.DL · cs.IR · cs.LG

    Identifying and assigning keywords at scale is a technical, practical, and ethical challenge for crowdsourced collections. This article reports the findings of the "Extracting Keywords from Crowdsourced Collections" project, which used the Their Finest Hour Online Archive, a crowdsourced Second World War digital collection hosted by the University of Oxford, as a case study. The project evaluated three Natural Language Processing approaches...

    arxiv.org/abs/2607.09324 · PDF

  9. 09

    Automatic Thematic Indexing of Large Literary Corpora: A Machine Learning Approach to Voltaire's Complete Works

    Miguel Arana-Catania, Gillian Pink, Glenn Roe

    cs.CL · cs.AI · cs.DL · cs.IR · cs.LG

    Thematic indexing -- the practice of assigning structured conceptual labels to sections of text -- is essential to scholarly access in large-scale literary and historical editions, yet it remains a largely manual, labour-intensive process. This paper explores the application of machine learning to automatic thematic indexing, using two substantial sub-corpora of the Complete Works of Voltaire as a test case: the Essai sur les mœurs et...

    arxiv.org/abs/2607.09316 · PDF

  10. 10

    Creativity, honesty and designed forgetting emerge in small hyperbolic language models

    Kwan Soo Shin, In Seok Kang, Yunkyung Min

    cs.CL · cs.AI · cs.HC · cs.LG

    Language models are optimised for scale, yet remain functional rather than companionable, and as an assistant personalises into a companion, accumulating memory of one user, it quietly becomes someone, and can silently acquire traits that harm that user. What a companion is becoming, and what would make it worth becoming, has no reliable instrument: trained human raters cannot agree on the answer (Fleiss kappa = 0.074). Here we show that...

    arxiv.org/abs/2607.09306 · PDF

  11. 11

    Complexity-Guided Component-wise Initialization for Language Model Pretraining

    Konstantin Garbers, Nicholas Oh

    cs.CL · cs.LG

    Pretrained language models often exhibit structured weight spectra, suggesting that training may repeatedly produce similar layerwise and component-wise organization. We ask whether these recurring spectral patterns can be reused as an initialization signal for GPT-2-style language-model pretraining. First, we analyze eleven pretrained GPT-2-style checkpoints that vary in size, language, tokenizer, and training corpus, measuring Frobenius...

    arxiv.org/abs/2607.09204 · PDF

  12. 12

    Augmenting Fundamental Analysis with Large Language Models: A RAG-Based System for Generating Investor Briefs

    Bartosz Ziółko, Kacper Dobrzeniewski

    cs.CL · cs.AI · q-fin.PM · q-fin.TR

    In this study, we examine the opportunities brought by Large Language Models (LLMs) to various aspects of fundamental analysis of companies based on their reports as well as data and documents describing macroeconomic situation like GDP and inflation changes as well as documents filled to the U.S. Securities and Exchange Commission (SEC) which can be found in EDGAR. We were preprocessing those data and than sending via API to gpt-4o model in...

    arxiv.org/abs/2607.09121 · PDF

  13. 13

    PRecG: Legal Precedent Retrieval with Graph Neural Networks and Rhetorical Role Segmentation

    Devanshu Verma, Vasudha Bhatnagar, Vikas Kumar, Balaji Ganesan

    cs.CL · cs.AI

    Legal precedent retrieval is a fundamental task in legal case preparation, planning, litigation strategy, and legal research. Current approaches for automatic precedent retrieval map legal documents to a low-dimensional semantic space and compute similarity based on the proximity of their representations. These approaches treat legal documents as monolithic texts, ignoring the rhetorical organization of the legal technicalities. Ergo, they...

    arxiv.org/abs/2607.09094 · PDF

  14. 14

    Validity of LLMs as data annotators: AMALIA on authority

    Manuel Pita

    cs.CL · cs.AI · cs.CY

    A national language model offers a linguistic community its own instrument for measuring what its citizens say and value. Portugal's AMALIA, a publicly funded 9B-parameter model for European Portuguese, appears competitive on agreement alone: asked to code the moral foundation of authority, it agrees with trained human coders to within six F1 points of open models eight to thirteen times its size. Yet agreement is reliability, not validity....

    arxiv.org/abs/2607.08731 · 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.