cs.CL · 2026-08-19 · No. 89

Computation and Language, 2026-08-19.

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

01 — The papers

12 entries
  1. 01

    TokEval: A Tokenizer Evaluation Suite

    Clara Meister

    cs.CL · cs.LG

    Language model tokenizers are typically selected with minimal evaluation, despite the fact that their design choices directly impact model capabilities. This can be partly attributed to a limited understanding of which tokenizer properties affect which aspects of downstream performance. We introduce TokEval, a framework of tokenizer evaluation metrics that goes beyond standard measures like fertility and compression rate to capture...

    arxiv.org/abs/2608.18062 · PDF

  2. 02

    Grading Needs a Rubric, Not Intelligence

    Jhen-Ke Lin

    cs.CL · cs.AI

    Small language models can grade open-ended examination answers as reliably as substantially more expensive models when they grade against an explicit rubric. We test this claim as the design principle behind any-to-bench: a frontier model reads source documents once, at ingestion, to extract each question and its rubric; lower-cost models then perform all repeated grading work. We evaluate six cost-efficient model configurations from two...

    arxiv.org/abs/2608.17938 · PDF

  3. 03

    BEAR-Bench: A Bilingual Enterprise and Academic Reasoning Benchmark for Multimodal Models

    Liubov Chubarova, Alexandra Kuleshova, Daniil Volkov, Kirill Sultanov, Alexey Zaytsev

    cs.CL · cs.AI

    While Multimodal Large Language Models (MLLMs) have made significant strides in visual comprehension, their ability to reason about text-dense, professional documents remains incompletely evaluated. Existing benchmarks emphasize information extraction, require external domain knowledge, or cover professional documents only as one of many settings. They are also largely English- or Chinese-centric, leaving other languages and Russian, in...

    arxiv.org/abs/2608.17895 · PDF

  4. 04

    Encoded but Not Actionable: Auditing the Decode-Generate-Steer Gap in Frozen LLMs for Geometric Constraints

    Man Liang, Xinzhao Cheng, Faizan Wajid

    cs.CL · cs.AI

    Large language models (LLMs) have demonstrated strong performance on structured reasoning tasks, but what they encode and whether it informs model behavior remain unclear. We investigate this question through geometric reasoning, using parametric CAD constraints as a controlled testbed for separating local pairwise relations from sketch-level constraint status. By probing the hidden states of six frozen decoder-only LLMs, we examine four...

    arxiv.org/abs/2608.17843 · PDF

  5. 05

    Interpretable Humans, Alien LLMs: Expert Analysis of Latent Structures in Assessment Responses

    Alona Strugatski, Licol Zeinfeld, Jason Cooper, Shelley Rap, Gil Schwarts, Giora Alexandron

    cs.CL · cs.AI · cs.HC

    The evaluation of large language models (LLMs) relies heavily on human-designed assessments, implicitly assuming that AI and humans employ similar underlying cognitive constructs. Challenging this assumption, we investigate whether the latent factors governing LLM performance carry the same substantive, human-interpretable meaning as the cognitive constructs governing human learners. Using responses from humans and six LLMs across...

    arxiv.org/abs/2608.17810 · PDF

  6. 06

    Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See

    Ayoub Kirouane, Christos Petrocheilos

    cs.CL · cs.LG · cs.RO · stat.ML

    Take three frontier mixture-of-experts models (Alibaba, OpenAI, NVIDIA; 3.6-4.0B active parameters each) and fine-tune them to reason in a low-resource language. On accuracy benchmarks almost nothing happens, and the benchmark itself is noise at this scale: changing only the random seed moves the score by 7.7 points, more than every data and recipe effect we measured. That null is our first result. The real changes live where accuracy cannot...

    arxiv.org/abs/2608.17744 · PDF

  7. 07

    Multi-turn Conversational AI from Text to Multimodal Interaction: Data, Models, Evaluation, and Open Challenges

    Syeda Faiza Ahmed, Zien Sheikh Ali, Hunzalah Hassan Bhatti, Firoj Alam, Shammur Absar Chowdhury

    cs.CL · cs.AI · cs.SD

    Conversational AI is moving beyond isolated text prompts toward sustained, multimodal interaction. In real conversations, users clarify goals, revise requests, interrupt responses, switch topics, and introduce new evidence while expecting systems to preserve context across turns. This makes multi-turn dialogue a distinct challenge requiring systems to maintain and update memory, ground responses across modalities, tools, and external...

    arxiv.org/abs/2608.17605 · PDF

  8. 08

    CoAL-RAG: A Complexity-Aware Legal Retrieval-Augmented Generation Method

    Jin Su, Zhuofeng Zhao, Huanhuan Wang, Hao Chen

    cs.CL · cs.AI

    Legal consultation questions exhibit multi-level complexity. A single retrieval strategy often leads to over-reasoning for simple questions and poor interpretability for complex ones, making it difficult to meet the requirements for both answer quality and efficiency in high-risk scenarios. To address this issue, this paper proposes CoAL-RAG, a complexity-aware legal retrieval-augmented generation method, which constructs a multi-dimensional...

    arxiv.org/abs/2608.17536 · PDF

  9. 09

    PTXBench: Benchmark and Adapt LLMs for GPU Kernel Optimization with Architecture-specific PTX

    Genghan Zhang, Yixin Dong, Chengze Fan, Zhichen Zeng, Yueming Yuan, Shaowei Zhu, Kunle Olukotun

    cs.CL · cs.AI

    We introduce PTXBench, a benchmark for evaluating and adapting large language models (LLMs) to use architecture-specific PTX for GPU kernel optimization. PTXBench measures functional correctness, whether selected target instructions execute at runtime, and speedup over frontier libraries across GEMM and attention workloads on H100 and B200 GPUs. Our evaluation shows that architecture-specific PTX capability remains uneven: success rates fall...

    arxiv.org/abs/2608.17379 · PDF

  10. 10

    Temporal Leakage in Financial News NLP: A Multi-Architecture Audit with a Regime-Specific M&A Signal

    Chenhao Xue, Raslen Guesmi, Siwei Feng, Yucheng Gong, Jacob Xavier Sundram, Jordan Pang, Lan Wang, Julian Kaljuvee

    cs.CL · cs.LG

    Financial-news direction prediction has become a popular NLP benchmark, yet reported gains depend critically on whether the train-test split is chronological or random, i.e., on temporal leakage. We audit this dependence on a 49,799-article corpus across 16 feature-model combinations spanning TF-IDF, MiniLM, FinBERT, and fine-tuned RoBERTa-large / DeBERTa-v3-large, plus separate zero/few-shot and LoRA probes of Llama-3 and Qwen2.5 LLMs:...

    arxiv.org/abs/2608.17223 · PDF

  11. 11

    Token Optimization and Context Window Management in Multi-Agent AI Workflows

    Dvir Shamay

    cs.CL · cs.AI

    Multi-agent AI workflows are limited not only by model quality but by token cost, latency, and context-window quality. This paper presents a practitioner framework for token optimization and context-window management, grounded in an internal production dashboard that extracts structured work items from meetings, email, and chat with LLMs and routes summaries across workstreams. Six patterns are described: context stratification,...

    arxiv.org/abs/2608.17188 · PDF

  12. 12

    Can LLMs Reason in a Legally Meaningful Manner? A Small-scale Study on European Court of Human Rights Cases

    Amogh Raina, Ilias Chalkidis, Daniel Hershcovich, Henrik Palmer Olsen

    cs.CL · cs.AI

    Reasoning has become a standard technique and feature for contemporary LLMs; however, its application and quality in the context of demanding legal-oriented tasks, such as legal case forecasting, remain under explored. We investigate how LLMs reason in the context of legal case forecasting, using legal cases from the European Court of Human Rights (ECtHR) as a testbed. We evaluate OpenAI GPT 5.4, a recent top-tier LLM, by exploring...

    arxiv.org/abs/2608.17168 · 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.