9 papers
Learning from Teacher Continuations at Student States
Haojin Wang, Dylan Zhang, Huaibo Chen +8
We present OLIVE (OnLine InterVEntion). At each iteration, the evolving student policy generates a new prefix, the teacher continues it autoregressively, and the student is updated…
How Can Rhetoric Reward-Hack AI Reviewers? Dissecting Rhetorical Sensitivity in AI-Based Peer Review
Ming Li, Chenguang Wang, Xirui Li +5
As large language models increasingly participate in scientific evaluation, we investigate a potential form of reward hacking: how rhetorical choices shape AI-review judgments when…
Sandboxed Coding Agents are Competitive Omni-modal Task Solvers
Dongping Chen, Xuanao Huang, Zhihan Hu +3
As multimodal LLMs increasingly target video and audio, it is often assumed that such tasks require native omnimodal models. We show that this is not always the case: coding agents…
Useful Memories Become Faulty When Continuously Updated by LLMs
Dylan Zhang, Yanshan Lin, Zhengkun Wu +4
Learning from past experience benefits from two complementary forms of memory: episodic traces -- raw trajectories of what happened -- and consolidated abstractions distilled acros…
Superminds Test: Actively Evaluating Collective Intelligence of Agent Society via Probing Agents
Xirui Li, Ming Li, Yunze Xiao +4
Collective intelligence refers to the ability of a group to achieve outcomes beyond what any individual member can accomplish alone. As large language model agents scale to populat…
What makes Reasoning Models Different? Follow the Reasoning Leader for Efficient Decoding
Ming Li, Zhengyuan Yang, Xiyao Wang +4
Large reasoning models (LRMs) achieve strong reasoning performance by emitting long chains of thought. Yet, these verbose traces slow down inference and often drift into unnecessar…
Florence-VL: Enhancing Vision-Language Models with Generative Vision Encoder and Depth-Breadth Fusion
Jiuhai Chen, Jianwei Yang, Haiping Wu +4
We present Florence-VL, a new family of multimodal large language models (MLLMs) with enriched visual representations produced by Florence-2, a generative vision foundation model.…
RuleR: Improving LLM Controllability by Rule-based Data Recycling
Ming Li, Han Chen, Chenguang Wang +3
Large language models (LLMs) still lack delicate controllability over their responses, which is critical to enhancing their performance and the user experience. However, curating s…
AutoHallusion: Automatic Generation of Hallucination Benchmarks for Vision-Language Models
Xiyang Wu, Tianrui Guan, Dianqi Li +9
Large vision-language models (LVLMs) are prone to hallucinations, where certain contextual cues in an image can trigger the language module to produce overconfident and incorrect r…