9 papers
PASTABench: Proactive Assessment of Sequential Trajectories for Agent Safety
Jiapeng Sun, Yujin Zhou, Han Zhu +4
As Large Language Models (LLMs) evolve into autonomous agents that alter real-world states, ensuring operational safety across multi-step workflows has become a critical challenge.…
Not Just the Destination, But the Journey: Reasoning Traces Causally Shape Generalization Behaviors
Pengcheng Wen, Yanxu Zhu, Jiapeng Sun +5
Chain-of-Thought (CoT) is often viewed as a window into LLM decision-making, yet recent work suggests it may function merely as post-hoc rationalization. This raises a critical ali…
What, Whether and How? Unveiling Process Reward Models for Thinking with Images Reasoning
Yujin Zhou, Pengcheng Wen, Jiale Chen +6
The rapid advancement of Large Vision Language Models (LVLMs) has demonstrated excellent abilities in various visual tasks. Building upon these developments, the thinking with imag…
Glance-or-Gaze: Incentivizing LMMs to Adaptively Focus Search via Reinforcement Learning
Hongbo Bai, Yujin Zhou, Yile Wu +5
Large Multimodal Models (LMMs) have achieved remarkable success in visual understanding, yet they struggle with knowledge-intensive queries involving long-tail entities or evolving…
Towards Multi-modal Multi-turn Safety: From Agentic Interaction to Strategic Alignment
Han Zhu, Jiale Chen, Chengkun Cai +8
Despite remarkable capability in multi-modal understanding, deploying Multi-modal Large Language Models (MLLMs) in open-ended conversational scenarios introduces safety risks that…
AI Deception: Risks, Dynamics, and Controls
Boyuan Chen, Sitong Fang, Jiaming Ji +56
As intelligence increases, so does its shadow. AI deception, in which systems induce false beliefs to secure self-beneficial outcomes, has evolved from a speculative concern to an…
SafeMT: Multi-turn Safety for Multimodal Language Models
Han Zhu, Juntao Dai, Jiaming Ji +8
With the widespread use of multi-modal Large Language models (MLLMs), safety issues have become a growing concern. Multi-turn dialogues, which are more common in everyday interacti…
J1: Exploring Simple Test-Time Scaling for LLM-as-a-Judge
Chi-Min Chan, Chunpu Xu, Jiaming Ji +7
The current focus of AI research is shifting from emphasizing model training towards enhancing evaluation quality, a transition that is crucial for driving further advancements in…
ThinkPatterns-21k: A Systematic Study on the Impact of Thinking Patterns in LLMs
Pengcheng Wen, Jiaming Ji, Chi-Min Chan +5
Large language models (LLMs) have demonstrated enhanced performance through the \textit{Thinking then Responding} paradigm, where models generate internal thoughts before final res…