Prépublication
· 2026 arXiv
EN
Expanding multilingual ASR systems to new languages is challenging when adaptation must preserve previously supported languages without replaying source-language training data. We study this problem in connector-based Speech LLMs, where a trainable connector m...
Yuchen Zhang, Haralambos Mouratidis et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
Human oversight is a pillar of AI safety. It is often exercised by a human in the loop (HITL) who approves critical actions taken by an AI agent. AI agents' increasing ability to bypass guardrails, though, raises the question of how to enforce HITL robustly. W...
Samuel Breckenridge, Dani Vilardell et al.
#cs.CR
Prépublication
· 2026 arXiv
EN
Frontier multimodal large language models (MLLMs) can control robots without robot-specific training, but how their capabilities compare and interact with post-trained vision-language-action (VLA) and world action model (WAM) policies across manipulation deman...
Yifei Yao, Yude Zou et al.
#cs.RO
Prépublication
· 2026 arXiv
EN
A vulnerability costs more to fix the later it is discovered. As coding agents generate code faster than developers can review it, software needs a first line of defense screening each new function in real time. LLM agents have recently become capable of findi...
Amine Lbath, Massih-Reza Amini
#cs.CR
Prépublication
· 2026 arXiv
EN
Selection bias in large language models (LLMs) can persist despite instructions to ignore presentation order or prompts that elicit reasoning. Recent efforts to reproduce Jev-style System One behavior have produced a growing family of models that predict proba...
Mingyang Song, Mao Zheng et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
A robot policy needs to understand what a task is asking for in a scene, and how to move the robot to accomplish that task. To achieve these two skills, we introduce Zero-real Agent Policy (ZAP), a recipe that teaches a small off-the-shelf vision-language mode...
Benjamin Pikus, Daniel Saunders et al.
#cs.RO
Prépublication
· 2026 arXiv
EN
Diffusion large language models (dLLMs) can generate multiple tokens in parallel, but often require many decoding rounds to complete a response. Efficient generation depends not only on which tokens are easy to predict, but also on how committing to them helps...
Aidar Myrzakhan, Zhiqiang Shen
#cs.CL
Prépublication
· 2026 arXiv
EN
Many academic disciplines (in particular, mathematics and the theoretical sciences) face difficult questions about how to adapt their practices to the rise of highly capable artificial intelligence (AI). Before AI, a well-written research paper with a strong t...
Saptarshi Pal, Peter S. Park
#cs.AI
Prépublication
· 2026 arXiv
EN
Deep research, or any use of large language models (LLMs) to read a large document collection and produce insights, requires synthesizing information from different sources. In this work, we find that LLMs produce long-form research reports that read as confid...
Tiasa Singha Roy, Manya Wadhwa et al.
#cs.CL
Prépublication
· 2026 arXiv
EN
Cybersecurity governance, risk management, compliance, and security architecture rely on numerous concepts, including security properties, domains, controls, risks, implementation measures, capabilities, and compliance requirements. Although these concepts are...
Hanane Houmani
#cs.CR
Prépublication
· 2026 arXiv
EN
Interactive environments offer a practical way to evaluate large language model (LLM) agents, and competitive games extend this evaluation to decisions shaped by competing goals and other agents' choices. Yet existing evaluations confined to individual games o...
Fengrui Zhang, Xiang Li et al.
#cs.AI
Prépublication
· 2026 arXiv
EN
The small Solar System bodies (SSSBs) preserve records of the formation and evolution of the Solar System. In particular, the near-infrared wavelength region around 3 $μ$m provides crucial insights into the origin and transport of hydrated materials and organi...
Jin Beniyama, Eri Tatsumi et al.
#astro-ph.EP