cs.CL · 2026-06-09 · No. 18
Computation and Language, 2026-06-09.
13 new papers in cs.CL. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
13 entries-
01
Data Synthesis and Parameter-Efficient Fine-Tuning for Low-Resource NMT: A Case Study on Q'eqchi' Mayan
Alexander Chulzhanov, Soeren Eberhardt, Arjun Mukherjee
cs.CL · cs.AI · cs.LG
Neural machine translation for digitally low-resource Indigenous languages is often hindered by extreme data scarcity, prompting reliance on extractive web-scraping. To ensure data sovereignty, this study introduces a data synthesis methodology to bootstrap NMT models without scraping target-language parallel text. Focusing on Q'eqchi' Mayan, we transformed community-sourced dictionaries into a massive synthetic corpus, utilizing...
-
02
Learning to Attack and Defend: Adaptive Red Teaming of Language Models via GRPO
Blake Bullwinkel, Eugenia Kim, Amanda Minnich, Mark Russinovich
cs.CL · cs.AI · cs.LG
AI red teaming must continually adapt to evolving attackers and defenders. Reinforcement learning offers a promising approach to discovering novel attacks, and co-training methods can produce more robust defenders in tandem. Recent works have demonstrated the efficacy of attacker-defender co-training by applying PPO and DPO, but report that GRPO is unstable in this setting. We introduce AdvGRPO, a co-training framework that makes GRPO viable...
-
03
End-to-End Context Compression at Scale
Ang Li, Sean McLeish, Haozhe Chen, Nimit Kalra, Zaiqian Chen, Artem Gazizov, Venkata Anoop Suhas Kumar Morisetty,...
cs.CL · cs.AI · cs.LG
Long-context language model inference is bottlenecked by memory, as the KV cache grows with context length. Recent techniques to compress the KV cache fall short: they either degrade model quality substantially or require considerable time and compute to compress a single long prompt. Furthermore, many methods require the input to fit within the target model's context window, and are generally incompatible with modern production inference...
-
04
AGENTSERVESIM: A Hardware-aware Simulator for Multi-Turn LLM Agent Serving
Rakibul Hasan Rajib, Mengxin Zheng, Qian Lou
cs.CL · cs.AI
Multi-turn LLM agents interleave model calls with external tool invocations, shifting serving from stateless request processing to stateful program execution. Serving these workloads requires scheduling, KV-cache management, and routing policies that use program-level context, including turn dependencies, tool-induced gaps, and reusable KV state. Evaluating such policies directly on real systems is costly, since each design point may require...
-
05
Code Is More Than Text: Uncertainty Estimation for Code Generation
Yuling Shi, Caiqi Zhang, Yuexian Li, Haopeng Wang, Yeheng Chen, Nigel Collier, Xiaodong Gu
cs.CL · cs.LG · cs.SE
Large language models (LLMs) are increasingly deployed as code generators, where silently wrong programs pose real safety and reliability risks. Reliable uncertainty estimation (UE) is essential for selective prediction, human-in-the-loop review, and downstream agentic decisions. Yet most existing code UE methods are inherited from natural language (NL) generation and ignore properties that make code distinct. We argue that code differs from...
-
06
Emergence of Context Characteristics Sensitivity in Large Language Models
Nadya Yuki Wangsajaya, Haeun Yu, Isabelle Augenstein
cs.CL · cs.AI
During instruction fine-tuning (IFT), large language models (LLMs) learn to follow instructions by using the provided context to answer a query. While prior work has studied how context characteristics correlate with context usage by the LLM, this analysis has been limited to inference time, leaving open how these relationships are acquired in the first place. Here, we measure how models' sensitivity to such characteristics shifts across...
-
07
Memory Beyond Recall: A Dual-Process Cognitive Memory System for Self-Evolving LLM Agents
Tianxiang Fei, Mingyang Song, Mao Zheng, Xiang Yu
cs.CL · cs.AI
Long-term memory for an LLM agent is more than retrieving the right passage at the right time. Current memory systems collapse belief revision, causal coupling, and cross-domain abstraction into a single retrieval surface tuned for surface recall, and consequently struggle on implicit personalisation that requires reasoning over how a user has evolved. We propose DCPM, which reorganises agent memory along a cognitive capability hierarchy...
-
08
A Finetuned SpeechLLM for Joint Multi-Granular L2 Assessment and Natural-Language Rationales
Aditya Kamlesh Parikh, Cristian Tejedor-Garcia, Catia Cucchiarini, Helmer Strik
cs.CL · cs.AI
Automated L2 speech assessment can assign proficiency labels, but often lacks interpretability. We propose a rubric-guided SpeechLLM for multi-aspect, multi-granular assessment, trained with a hybrid objective combining supervised fine-tuning and Bounded Direct Preference Optimization. The model jointly predicts ordinal labels at the sentence-level (accuracy, fluency, prosody), word/phoneme-level accuracy, and generates a natural-language...
-
09
PriFT: Prior-Support Guided Supervised Fine-Tuning
Ke Wang, Shuangqi Li, Mathieu Salzmann, Pascal Frossard
cs.CL · cs.LG
Supervised fine-tuning (SFT) is an efficient approach for downstream task adaptation and often serves as the initialization stage for reinforcement learning (RL), but it can show weaker generalization than RL. A key limitation is its off-policy objective: SFT fits fixed demonstrations token by token, including targets poorly aligned with the model's pretrained distribution, which can lead to overfitting. A recent line of work addresses this...
-
10
SG-OPD: Sign-Gated On-Policy Distillation via Sign-Consistency Gating and Phased Teacher Sampling
Haoran Xu, Hongyu Wang, Yifei Gao, Jiaze Li, Xiaofeng Zhang, Xiaosong Yuan
cs.CL · cs.LG
On-policy distillation (OPD) trains a student on its own trajectories with dense per-token supervision from a stronger teacher, and often outperforms off-policy distillation and standard reinforcement learning. However, we find that its effectiveness implicitly relies on two assumptions that frequently break in practice: trajectory-level alignment between the student and the teacher, and uniform token-level reliability of the teacher's...
-
11
Culturally-Adapted Red-Teaming Across East and Southeast Asian Contexts: A Methodological and Comparative Analysis
Hyeji Choi, Yongtaek Lim, Minwoo Kim
cs.CL · cs.AI
Multilingual safety evaluation of large language models (LLMs) has predominantly relied on direct translation (DT) of English benchmarks into target languages - an approach that converts surface-level linguistic form while failing to reflect the cultural context embedded in threat scenarios, social norms, and legal frameworks. We construct paired DT and culturally-adapted (CA) datasets via 1:1 seed matching for four languages - Korean (KO),...
-
12
Unified Energy for Invariant and Independent Decoding in Diffusion Language Models
Yuchen Yan, Minkai Xu, Zaiquan Yang, Yatao Bian
cs.CL · cs.AI
Diffusion Language Models (DLMs) enable parallel text generation by iteratively denoising a full sequence, offering attractive flexibility compared to auto-regressive (AR) decoding. However, existing methods fail to fully capture token relationships, leading to a performance gap relative to AR baselines, especially as the degree of parallelism increases. In this paper, we give a systematic analysis of the gap, identifying three key factors:...
-
13
SEF-CLGC at SemEval-2026 Task 11: Logical Notation Impact on Language Model Performance
Hanna Abi Akl, Fabien Gandon, Catherine Faron, Pierre Monnin
cs.CL · cs.AI
This paper revisits our pipeline called Syllogistic Evaluation Framework-Common Logic Grammar Construction (SEF-CLGC). We combine formal logical notations with Small Language Models (SLMs) to evaluate reasoning performance on the SemEval-2026 Task 11 Subtask 1: Disentangling Content and Formal Reasoning in Large Language Models. Our experiments show that by relying solely on SLMs, trained on a combination of natural and symbolic languages,...
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.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.