cs.CL · 2026-08-08 · No. 78

Computation and Language, 2026-08-08.

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

01 — The papers

15 entries
  1. 01

    Learning When to Trust via Selective Context Preference Optimization

    Xian Sun, Wei Chow, Yingshuo Wang, Junhao Liu, Wei Gao, Qing Wu, Lingdong Kong

    cs.CL · cs.AI · cs.LG

    Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong. The obvious remedy, training models to resist such signals, hides a failure mode: a model that ignores all context looks robust yet is useless when the context is worth trusting. We recast the problem as selective trust and introduce MIST, a human-annotated benchmark that renders each reasoning item under four...

    arxiv.org/abs/2608.06377 · PDF

  2. 02

    Benchmarking the Benchmarks: Evaluating Benchmarks for Conversational Agents

    Noam Koren, Roy Bar-Haim, Abigail Goldsteen

    cs.CL · cs.AI

    Task-oriented conversational agents are evaluated using curated or automatically generated benchmarks, yet benchmark quality is rarely assessed. Poor benchmarks may contain inconsistent tasks, simplistic scenarios, or limited policy coverage, leading to unreliable evaluations. We introduce a reference-free framework that uses LLM judges to assess benchmark consistency, complexity, and policy coverage, while providing actionable diagnostics of...

    arxiv.org/abs/2608.06329 · PDF

  3. 03

    Beyond Sequence Order: Syntax-Informed Positional Embeddings for Transformers

    Haris Riaz, Hyungji Kim, Mihai Surdeanu

    cs.CL · cs.AI

    Positional embeddings (PE) in Transformers encode token distance and order but are largely agnostic to \textit{syntactic structure}. We introduce \textbf{S}yntax-\textbf{i}nformed \textbf{P}ositional \textbf{E}mbeddings (\textbf{SiPE}), which learns a lightweight syntactic prior from dependency parses during pretraining and injects it across all three dominant PE families (absolute, relative, rotary), for both encoders and decoders, leaving...

    arxiv.org/abs/2608.06111 · PDF

  4. 04

    FormBharo: Designing and Evaluating a Voice Agent for Conversational Form Filling in Rural India

    Aman Dalmia, Sanskriti Midha, Jigar Doshi

    cs.CL · cs.AI · cs.HC

    In India, almost every social benefit starts with a form, yet the people who need these benefits most are often unable to read or write. Reaching them requires a spoken conversation. Today that work falls to frontline health workers who enroll beneficiaries one at a time, a poor use of stretched capacity. We built FormBharo ("fill the form" in Hindi), a voice agent that fills a structured form over a phone call under tight latency and cost...

    arxiv.org/abs/2608.06027 · PDF

  5. 05

    MACRO: Markov Chain Routing of Transformer Layers

    Paweł Batorski, Abtin Pourhadi, Akylgali Aitaza, Przemysław Spurek, Paul Swoboda

    cs.CL · cs.AI

    Standard Large Language Models (LLMs) execute layers sequentially. Dynamic layer routing, i.e. search for a different execution path through layers involving layer repetitions, skips and other moves, can improve performance. Existing routing approaches often require updating model weights, running expensive search loops per test instance, or demand ground-truth labels during inference. In this work, we propose Markov Chain Routing of...

    arxiv.org/abs/2608.05872 · PDF

  6. 06

    MameLoshnLM: Yiddish Language Model and Evaluation Benchmark

    Uri Katz, Omer Goldman, Tomasz Limisiewicz, Reut Tsarfaty, Noah A. Smith

    cs.CL · cs.AI

    We present MameLoshnLM, the first open-source 8B-parameter language model built specifically for Yiddish. Despite Yiddish's rich textual tradition, its limited digital presence and the scarcity of reliable evaluation resources have constrained progress in Yiddish language modeling. Existing multilingual corpora and benchmarks are often poor proxies for the language, containing substantial amounts of noisy, machine-translated, and...

    arxiv.org/abs/2608.05850 · PDF

  7. 07

    Hierarchical Latent Prediction for Language Models

    Chang Shi, Tim Pearce, Manan Tomar, Siddhartha Sen, John Langford

    cs.CL · cs.AI

    While standard Next-Token Prediction (NTP) lays the foundation of language model pre- training, its teacher-forced training paradigm may not be optimal for long-horizon reasoning and planning. Recent works such as Multi-Token Prediction (MTP) and Next-Latent prediction (NextLat) try to mitigate the problem through predicting multiple future tokens and self-supervised prediction in the latent space. However, those auxiliary objectives either...

    arxiv.org/abs/2608.05806 · PDF

  8. 08

    On-Policy Delta Distillation for Multilingual Math Reasoning

    Byeongho Heo, Jaehui Hwang, Sangdoo Yun, Dongyoon Han

    cs.CL · cs.LG

    On-Policy Distillation (OPD) is emerging as a promising alternative to reinforcement learning for LLM post-training, yet its effectiveness in multilingual settings remains underexplored. We study OPD and its advanced variant, On-Policy Delta Distillation (OPD$^2$), for mathematical reasoning in English, Korean, and Japanese. OPD$^2$ improves OPD by using the probability gap between a post-trained teacher and its base model as the learning...

    arxiv.org/abs/2608.05802 · PDF

  9. 09

    Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation

    Tirth Bhatt, Naren Kumar S, Mayank Singh

    cs.CL · cs.AI

    Multilingual text embedding models are commonly adapted using a single training objective across diverse tasks, despite different tasks requiring fundamentally different optimization strategies. We introduce Task-Conditional Flow Matching (TCFM), a multilingual embedding adaptation framework that selectively applies Flow Matching to translation tasks while optimizing retrieval, classification, and pair-classification tasks with objectives...

    arxiv.org/abs/2608.05785 · PDF

  10. 10

    Once a Response, Always a Response: Detecting LLM-generated Text via Latent Prompt Restoration

    Hongrui Bao, Yubing Ren, Yanan Cao, Jinhan You, Fang Fang, Shi Wang

    cs.CL · cs.AI

    Large language models (LLMs) can generate fluent and convincing text at scale, creating growing risks for misinformation dissemination, educational misuse, and platform governance. These concerns make robust detection of machine-generated text increasingly necessary. Recent zero-shot detectors mainly exploit probability-based statistical discrepancies, but they do not explicitly account for the training process of LLMs, which leaves a...

    arxiv.org/abs/2608.05741 · PDF

  11. 11

    Sparse Mutual Information Graph Averaging for Improving Random Indexing Embeddings

    Sriram Loganathan, Gokul Anand, Aung Bo Bo, Yourui Shao, William B. Andreopoulos

    cs.CL · cs.LG

    Sparse word embedding pipelines can avoid dense co-occurrence matrix materialization, dense factorization, and gradient training while still relying on sparse global corpus statistics. This paper studies Random Indexing (RI) vectors refined by weighted averaging on a sparse Positive Pointwise Mutual Information (PPMI) graph. On a fairytales corpus, the covered semantic analogy set consists of 272 Google family- category questions. On this...

    arxiv.org/abs/2608.05724 · PDF

  12. 12

    Answer First, Reason Later: Commitment Order in Diffusion LLMs

    Jewon Yeom, Jaewon Sok, Seonghyeon Park, Jeongjae Park, Hwiyeong Lee, Taesup Kim

    cs.CL · cs.AI

    Masked diffusion language models (dLLMs) can commit tokens in any order -- a freedom marketed as their core advantage over autoregressive decoding. We show that on reasoning tasks this freedom is instead the axis of failure. Logging every commitment during decoding of LLaDA-8B on GSM8K, we find that unconstrained (pure) decoding commits the final answer at 15-24% of the trajectory while half the reasoning region is still masked, and collapses...

    arxiv.org/abs/2608.05687 · PDF

  13. 13

    Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolution

    Sichun Luo, Yi Huang, Guanzhi Deng, Haibo Wang, Haochen Luo, Lei Li, Zefa Hu, Junlan Feng, Qi Liu

    cs.CL · cs.AI · cs.NE

    Large language model (LLM)-driven evolution has shown promise for program search and algorithm discovery, but relying on strong models throughout long evolutionary runs is costly. A natural alternative is to combine cheap and strong models under a fixed inference budget. However, existing approaches typically allocate models at the level of individual queries or mutation steps, overlooking that evolutionary search is \textit{stateful}: each...

    arxiv.org/abs/2608.05651 · PDF

  14. 14

    FOCUS: Decoupling Expert Personas in LLMs to Enhance Domain Expert Capabilities

    Guanyu Wang, Zidi Zhang, Xu Chu

    cs.CL · cs.LG

    Large Language Models (LLMs) can exhibit diverse personas, and activating expert personas has been shown to improve domain expertise and task accuracy. However, existing persona control methods often suffer from cross-domain coupling, which may lead to overly aggressive behavior in high-caution domains such as healthcare, or excessive conservatism in risk-sensitive domains such as financial trading. To address this issue, we propose FOCUS...

    arxiv.org/abs/2608.05611 · PDF

  15. 15

    SkillZip: Contract-Preserving Graph Compression for Scalable Agent Skill Libraries

    Xingyu Tan, Xiaoyang Wang, Qing Liu, Xiwei Xu, Xin Yuan, Liming Zhu, Wenjie Zhang

    cs.CL · cs.AI

    Large Language Models (LLMs) increasingly act as agents whose procedural knowledge is stored in reusable skill packages and loaded at inference time. As skill libraries grow, a central challenge is to expose the smallest sufficient executable context under a limited context budget. Existing systems struggle to reuse routines below the whole-skill level, preserve procedural contracts during compression, keep compressed routines executable and...

    arxiv.org/abs/2608.05604 · 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.