cs.CL · 2026-10-21 · No. 150

Computation and Language, 2026-10-21.

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

01 — The papers

17 entries
  1. 01

    Benchmarking Arabic Dialects for Math Problems

    Noor Abo Mokh, Teresa Lynn, Nizar Habash, Ahmed Elshabrawy, Sara Alhajeri, Alaa Elsetohy, Preslav Nakov, Alham Fikri Aji

    cs.CL · cs.AI

    Mathematical reasoning remains underexplored for low-resource languages, including Dialectal Arabic. To bridge this gap, we introduce the DAR-MGSM (Dialectal Arabic Multilingual Grade School Math) Benchmark, the first manually curated benchmark for evaluating mathematical reasoning in Dialectal Arabic. DAR-MGSM consists of math word problems (MWPs) requiring multi-step reasoning and covers five major Arabic dialects alongside Modern Standard...

    arxiv.org/abs/2610.28607 · PDF

  2. 02

    Capable Teacher, Myopic Scorer: Reward-Free Teacher Repair for Agentic On-Policy Distillation

    Shu Zhao, Tan Yu

    cs.CL · cs.AI

    Agentic on-policy distillation (OPD) trains a student agent on its own trajectories while a teacher scores every token. A natural explanation of where it goes wrong is compounding error: the student's mistakes carry its trajectories beyond the region where the teacher is reliable, so recent methods keep training inside that region, through episode-length curricula, teacher-guided turns, prefix replay, or teacher intervention where the two...

    arxiv.org/abs/2610.28586 · PDF

  3. 03

    Large Flow Language Models

    Arnav Chavan, Chanhyuk Lee, Udbhav Bamba, Jaehoon Yoo, Seunghoon Hong, Aditi Raghunathan, Jinwoo Kim

    cs.CL · cs.AI

    Continuous flow language models offer a path to parallel generation with tools for control and acceleration unavailable to discrete diffusion, yet effective techniques for building them at the billion-parameter scale remain less established. We present Lafa (Large flow language model), a fast, high-performing flow language model scaled to 8B parameters in a data-efficient manner. Lafa is a blockwise flow model with causal attention on the...

    arxiv.org/abs/2610.28559 · PDF

  4. 04

    BitMTP: When Multi-Token Prediction Meets Low-Bit Large Language Models

    Ning Zhang, Shihao Wang, Jinrui Zhang, Chaodong Xiao, Lei Zhang

    cs.CL · cs.AI · cs.DC · cs.LG

    For efficient deployment of large language models (LLMs), methods of ultra-low-bit backbones reduce weight storage, memory traffic, and linear-layer execution cost, while multi-token prediction (MTP) methods reduce sequential decoding steps by predicting multiple future tokens in parallel. Although complementary, the direct combination of these methods cannot accelerate decoding at comparable quality: masked drafting lacks intermediate token...

    arxiv.org/abs/2610.28500 · PDF

  5. 05

    Every Model Leaves a Trace: Multi-Resolution LLM Fingerprinting

    Mohammad Reza Modarres, Armin Tourajmehr, Artin Tavasoli, Yadollah Yaghoobzadeh, Mohammad Taher Pilehvar

    cs.CL · cs.AI

    Large language models leave model-specific traces in their generated text, but it remains unclear how finely these fingerprints can distinguish related models and which signals remain informative as obvious cues are reduced. We study this through fingerprint resolution, considering both model resolution, from different families to distilled and adapted variants, and signal resolution across behavioral, stylometric, lexical, and semantic...

    arxiv.org/abs/2610.28434 · PDF

  6. 06

    The Instruction Tax on Factual Recall: Base Models Can Beat Their Instruction-Tuned Counterparts

    Christophe Khalil, Aya Mourad

    cs.CL · cs.AI

    We present our solution to the AKBC LM-KBC 2026 Shared Task, where we achieved a score of 0.6836 on the official leaderboard, more than double the baseline of 0.2964. This task requires generating all possible objects for subject-relation pairs using a language model with up to 32 billion parameters, with no retrieval or fine-tuning involved. We show that the prompting methodology is as important as the model architecture: while the...

    arxiv.org/abs/2610.28407 · PDF

  7. 07

    Beyond a Single Sentiment Score: Topical and Affective Dimensions of Emoji Meaning on Twitter

    Rohan James, Ioannis Christofilogiannis, Yuuv Jauhari, Shreya Shah, Gauri Pethkar

    cs.CL · cs.AI · cs.SI

    Emoji are widely treated as compact affective tokens whose meaning is summarised by a single sentiment score. We argue that this picture is incomplete. From a corpus of 16{,}575 English tweets and a per-emoji metric we call \emph{stability delta} (within-class minus between-class cosine similarity of sentence embeddings), we show that emojis used in tightly clustered topical contexts also exhibit \emph{more} polarised crowd-sourced sentiment,...

    arxiv.org/abs/2610.28337 · PDF

  8. 08

    Adaptive Replay Refinement: Allocating Test-Time Compute Between Revision and Resampling

    Ahsan Bilal, Muhammad Ahmed Mohsin, Muhammad Umer, Ayesha Mohsin, Sarmad Bilal, Dean F. Hougen

    cs.CL · cs.LG

    Test-time scaling allocates additional inference compute to refining existing solutions or generating new ones. This raises a fundamental question: Should I keep improving this solution, or stop working on it and try a completely new solution? Existing methods make this decision using fixed refinement schedules, self-verification judgments, or learned controllers, which can waste compute or rely on signals that may be incorrect or require...

    arxiv.org/abs/2610.28294 · PDF

  9. 09

    Temporal Evidence-Parameter Co-Alignment for LLM Reasoning over Temporal Knowledge Graphs

    Jiapu Wang

    cs.CL · cs.AI

    Large Language Models (LLMs) often struggle with temporally grounded reasoning because their static parametric memory is misaligned with the dynamic evolution of real-world knowledge. Although RAG and Parameter-Efficient Fine-Tuning (PEFT) provide practical ways to enhance LLMs, they remain insufficient for temporal reasoning. Existing retrieval methods typically prioritize semantic relevance while overlooking temporal validity, event order,...

    arxiv.org/abs/2610.28176 · PDF

  10. 10

    Africa-Papers: A 10-Billion-Token Multi-Domain Academic Corpus for African Foundation Language Models

    Mich-Seth Owusu

    cs.CL · cs.AI

    Large Language Models (LLMs) depend fundamentally on the diversity, depth, and factual rigor of their pre-training corpora. However, contemporary foundational pre-training datasets (such as The Pile, S2ORC, and Dolma) exhibit a severe structural omission: African scientific, medical, legal, and agricultural scholarship constitutes less than 0.5% of global pre-training tokens. Consequently, state-of-the-art models suffer from systematic...

    arxiv.org/abs/2610.28173 · PDF

  11. 11

    BAIT: Can LLMs Find What Biases a Reader? A Benchmark for Cognitive Bias Triggers in News

    Yousra Fettach, Tijl De Bie

    cs.CL · cs.AI

    Persuasive writing exploits cognitive biases, recruiting a reader's mental shortcuts to shape belief, often without the reader noticing. We call the linguistic features exploiting such mental shortcuts cognitive bias triggers. As large language models (LLMs) increasingly mediate how people encounter news, summarizing it, ranking it, and answering questions over it, a natural question is whether they can recognize the features of a text that...

    arxiv.org/abs/2610.28092 · PDF

  12. 12

    Same Judgment, Different Yes: Measuring Multilingual Judgment Profiles in Language Models

    Haonan Huang, Yuwei Lu, Chengkun Cai

    cs.CL · cs.AI

    Multilingual evaluations often score a model's opening yes or no, a token that also carries model and language answering conventions. We ask when different ways of eliciting and reading an answer recover the same judgment, and when they measure an answer habit. We collect 1.1 million replies from eight models in six languages on up to 105 moral vignettes, varying the question's form, the answer constraint, and the reading of the reply, token...

    arxiv.org/abs/2610.28066 · PDF

  13. 13

    On-Policy Transition Distillation

    Runyang You, Yongqi Li, Tiezheng Yu, Wenjie Li

    cs.CL · cs.AI · cs.LG

    On-policy distillation (OPD) consolidates expert capabilities into a single student by matching teacher and student final-layer distributions on student-generated rollouts. We investigate whether depthwise prediction transitions, the changes in predictions between layers, can provide additional task-relevant supervision. Our diagnostic study finds that stronger teacher--student agreement in intermediate transitions is associated with higher...

    arxiv.org/abs/2610.28048 · PDF

  14. 14

    Polarity Blindness in LLM Safety: A Systematic Study of Negation, Antonym Substitution, and Adversarial Framing

    Diego Miguel Lozano, Erik Derner, Juan Antonio Pérez-Ortiz, Nuria Oliver

    cs.CL · cs.AI

    Negation remains a persistent challenge for language models across NLP tasks and modalities. We show that this weakness can undermine LLM safety guardrails through a structural vulnerability we term polarity blindness: the failure to recognize harmful intent after polarity-altering linguistic transformations. We study two forms of polarity inversion: grammatical negation and antonym replacement, and use synonym replacement as a...

    arxiv.org/abs/2610.27908 · PDF

  15. 15

    CollabSciReM: Aligning Language Models for Scientific Conversations via Science-Aware Reward Decomposition

    Sasi Kiran Gaddipati, Gollam Rabby, Sören Auer

    cs.CL · cs.AI

    Large Language Models (LLMs) are increasingly used as research collaborators for hypothesis refinement, methodological planning, and technical analysis across multi-turn interactions. In this setting, a useful assistant must be accurate, methodologically justified, logically consistent, and token-efficient. Current alignment approaches rarely achieve all of these together. Instruction-tuned models gain conversational quality through...

    arxiv.org/abs/2610.27901 · PDF

  16. 16

    SMART: Semantic Segment Mapping and Assessment of Recognition Transcripts with Contextual Rubrics

    Xiaoyu Gu, Chencheng Wang, Fuyang Li, Yu Xi, Tao Liu, Jiaoqi Guo, Kai Yu

    cs.CL · cs.AI · eess.AS

    Word error rate (WER) assigns equal cost to lexical edits despite their different semantic consequences, while overall semantic scores cannot identify which individual errors alter the intended meaning. We formulate ASR semantic evaluation as context-aware, segment-level consequence assessment and propose Semantic Segment Mapping and Assessment of Recognition Transcripts with Contextual Rubrics (SMART), a framework for traceable semantic...

    arxiv.org/abs/2610.27820 · PDF

  17. 17

    ProVe: Proposer-Verifier Decoding

    Ziqi Jin, Weiwen Xu, Jiaxi Li, Yujie Zhong, Qi She, Aixin Sun

    cs.CL · cs.LG

    Speculative decoding accelerates autoregressive (AR) language models by letting a small model draft tokens serially and a large model verify them in parallel. We reverse these roles. Starting from a standard AR checkpoint, we fine-tune the large model into a proposer that predicts a block of future tokens in a single forward pass, then use one additional forward pass to verify the proposed block. This gives ProVe, Proposer-Verifier decoding....

    arxiv.org/abs/2610.27815 · 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.