cs.CL · 2026-11-08 · No. 167

Computation and Language, 2026-11-08.

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

    Label-Wording Augmentation in a Typed-Decision Encoder: Gains on Unfamiliar Words, Not on Abbreviations or Codes

    Vladimir Beskorovainyi

    cs.CL · cs.AI · cs.LG

    Typed-decision models score options that the caller supplies at request time, so their accuracy depends on label wording that the model's developer does not control. We compare two releases of a 307M-parameter multilingual cross-encoder, Ru-Jev, trained on thirteen Russian-language tasks; the later release adds an augmentation that presents labels in several surface forms. Holding questions fixed and varying only how eighteen routing labels...

    arxiv.org/abs/2611.08439 · PDF

  2. 02

    Understanding AI from the Inside: The State of Mechanistic Interpretability for AI Safety

    Fazl Barez, Tung-Yu Wu, Hunar Batra, Gabriele Dominici, Florent Draye, Yanai Elazar, Damiano Fornasiere, Antía...

    cs.CL · cs.AI · cs.LG

    Neural networks are likely to be among the most consequential artefacts in human history. As these systems become more capable and autonomous, their potential to cause societal harm also grows, making AI safety one of the defining challenges of our time. Yet we continue to build increasingly powerful systems without fully understanding how they work, how they will behave in unfamiliar situations, or how to reliably control them. Mechanistic...

    arxiv.org/abs/2611.08438 · PDF

  3. 03

    MedGraph: Maintaining Patient State with Graph Memory for Long-Term Medical Dialogue

    Donghua Zhang, Zizheng Wang, Ze Dong, Yiding Feng, Siyu Qian

    cs.CL · cs.AI

    LLM-based medical assistants that follow a patient across visits need a memory that keeps the patient's clinical state correct as it changes. We examine the errors of five memory systems on longitudinal medical dialogue and find three recurring causes: the retrieved evidence gives an outdated state, includes another person's health facts, or lacks a safety fact the answer depends on. We present MedGraph, a graph memory built for these three...

    arxiv.org/abs/2611.08433 · PDF

  4. 04

    Measured Coupling for Parallel Decoding in Diffusion Language Models

    Tal Oved, Lior Wolf

    cs.CL · cs.AI · cs.DC · cs.LG · cs.PF

    Diffusion language models can write many tokens per step, but must choose which tokens are safe to commit together. Mean-field decoding penalizes pairs of positions whose predictions look alike. We replace this proxy with the coupling itself: how much committing one token changes the model's confidence at another. Cheap probes inside the main forward pass measure it, with no training or extra model, and it plugs into the same mean-field rule....

    arxiv.org/abs/2611.08307 · PDF

  5. 05

    Knowing When to Teach: Learning from Teachable Prefixes in Multilingual Reasoning Distillation

    Jaewon Cho, Minseo Yoon, Jinyoung Park, Jinwoo Seo, Jaewon Chu, Alice Oh, Hyunwoo J. Kim

    cs.CL · cs.AI

    On-Policy Self-Distillation (OPSD) improves reasoning by training a student on its own reasoning, using the same model conditioned on reference solutions as a teacher. Applying OPSD to multilingual reasoning requires the teacher to provide useful feedback across languages, even when the student makes mistakes in intermediate steps. We observe a phenomenon we call Teacher Drift: as these mistakes accumulate, the teacher may follow the...

    arxiv.org/abs/2611.08179 · PDF

  6. 06

    CardioLM: A multilingual suite of small language models for the cardiology domain

    Manuela-Daniela Danu, Miguel Rodríguez-Ortega, Alberto Becerra-Tome, Pablo Arancibia, Fernando Gallego-Donoso,...

    cs.CL · cs.AI

    Background: Cardiovascular care generates large volumes of clinical free-text documentation in local languages. However, clinical natural language processing (NLP) resources remain concentrated in English, and no existing benchmark is cardiology-specific, multilingual, and broad in entity coverage. Methods: We present two contributions. The first is CardioCCC, a corpus of 508 cardiology case reports originally published in Spanish and...

    arxiv.org/abs/2611.07940 · PDF

  7. 07

    CORE: Coarse-to-Fine Search with Context Management for Deep Research

    Hoang Tran Vuong, Ngoc Quan Pham, Hung Bui, Fatih Porikli, Dat Quoc Nguyen

    cs.CL · cs.AI · cs.IR

    Deep research agents answer complex questions through repeated search. They rely on a harness that runs their tool calls and returns evidence to their context. This creates a trade-off: Search works best over long passages, which give enough context for reliable semantic matching. But the returned evidence should be short, because every observation stays in the agent's context for the rest of the trajectory. We introduce CORE, a training-free...

    arxiv.org/abs/2611.07859 · PDF

  8. 08

    Axis-LoRA: Probing and Adapting Embedding Models for Personal-Memory Retrieval

    Dohoon Kim, Junho Park, Hyunwoong Bae, Seyoon Kim, Taesup Moon

    cs.CL · cs.AI

    As LLMs are increasingly asked questions about a user's personal experiences, retrieval must distinguish memories by time, place, person, and event. Current personal-memory pipelines rely on frozen dense embedding models to embed memories for retrieval, but we find that these models do not capture the four axes well. Swapping a memory's date or place barely shifts its embedding, and a trained projection on top of the frozen embedding does not...

    arxiv.org/abs/2611.07820 · PDF

  9. 09

    Decodability Is Not Validity: Evaluating Nuisance Sensitivity in Language Model Concept Probes

    Ayal Klein

    cs.CL · cs.LG

    Two concept probes can be equally accurate at predicting a concept from a language model's hidden states while their scores contain different mixtures of the target concept and a correlated nuisance attribute. In four target--nuisance settings (two aspect sentiments in CEBaB; occupation and gender in BiasBios; Amazon review sentiment with negation, and with first-person framing), two LLMs, and a population of concept extractors, we compare...

    arxiv.org/abs/2611.07805 · PDF

  10. 10

    Bridging Structure and Semantics: Knowledge Graph Verification via Triple-to-Text Alignment

    Ziang Cui

    cs.CL · cs.AI

    Semantic accuracy is an important dimension of knowledge graph quality. Graph-based scoring methods exploit relational structure, while textual representations allow classifiers to draw on knowledge encoded by pre-trained language models (PLMs). We present GenVer, a generative verifier that scores triples using both their serialized form and generated natural language representations (NLRs). A triple-text alignment procedure selects training...

    arxiv.org/abs/2611.07780 · PDF

  11. 11

    Extending Reasoning in Typed Decision Models via Looping

    David Berghaus, Arber Zela, Behzad Shomali, Markus Frey, Mehdi Ali, Rafet Sifa, Volkan Cevher

    cs.CL · cs.AI · cs.LG

    Typed decision models choose among a fixed set of answer options in a single forward pass, so unlike generative language models they cannot spend more test-time compute on hard inputs by generating more tokens. Looping a block of layers offers this compute in latent space without adding parameters, but prior work found no gain from loops retrofitted into conventionally pretrained models. We show that such loops help when trained together with...

    arxiv.org/abs/2611.07736 · PDF

  12. 12

    Evaluating Token Probabilities as Confidence Estimators in Human Behavior Simulations: A Case Study in Luxembourgish Next Like Prediction

    Nils Schwager, Simon Münker, Alistair Plum, Achim Rettinger

    cs.CL · cs.AI

    As large language models (LLMs) are increasingly used to simulate human behavior, their operational validity is typically assessed through accuracy alone, implicitly treating human behavior as deterministic. We argue that on noisy digital trace data, a model's performance depends not only on predicting the correct outcome but on its calibration. We present a case study testing whether the token probabilities of open-weight LLMs can serve as...

    arxiv.org/abs/2611.07735 · PDF

  13. 13

    O$^2$PD: Overlap-Gated On-Policy Distillation on Reliable Prefix

    Shuaishuai Zu, Jihao Zhao, Pengfei Li, Guanghuan Fang, Ben Feng, Yuhan Liu, Biao Qin, Baochuan Yang

    cs.CL · cs.AI

    On-policy distillation (OPD) supplies dense token-level supervision by aligning a student to the teacher's next-token distribution on its own generated rollouts. Existing works mostly implement top-$k$ OPD, applying reverse Kullback-Leibler ($\RKL$) divergence between the teacher's and the student's renormalized distributions over teacher top-$k$ support. However, the teacher scores student-generated rollouts that are inherently off-policy...

    arxiv.org/abs/2611.07731 · PDF

  14. 14

    Continued Training of Continuous Language Flows with Token Identity Squared Supervision

    Jiawei Jiang, Zhaoxiang Zang, Zhengtao Yu

    cs.CL · cs.AI

    Diffusion and flow models have recently been extended to language modeling, generating text through continuous denoising in token embedding spaces. Pretrained models provide token representations and contextual prediction capabilities, but improving generation with limited additional training requires an appropriate connection between supervision and continuous generation. Continuous MSE evaluates a predicted vector obtained by weighting...

    arxiv.org/abs/2611.07721 · PDF

  15. 15

    English Calibration Erodes Lithuanian in Expert-Pruned Mixture-of-Experts Language Models

    Tomas Rasymas, Povilas Kvedaras

    cs.CL · cs.LG

    Expert pruning compresses Mixture-of-Experts (MoE) language models by removing the experts that a calibration set marks as unimportant. Studies of high-resource languages largely find that English calibration transfers well to other languages. We test this for Lithuanian: we pruned four MoE models with REAP, removing 25% or 50% of the experts with English, Lithuanian or mixed calibration, and evaluated both languages on reading comprehension,...

    arxiv.org/abs/2611.07654 · 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.