Publications (9)
Fleming-VL: Towards Universal Medical Visual Reasoning with Multimodal LLMs
Yan Shu, Chi Liu, Robin Chen +2
Multimodal Large Language Models (MLLMs) have demonstrated remarkable effectiveness in various general-domain scenarios, such as visual question answering and image captioning. Rec…
UniReason-Med: A Shared Grounded Reasoning Interface for 2D-to-3D Transfer in Medical VQA
Mengzhuo Chen, Yan Shu, Chi Liu +4
We study whether grounded reasoning supervision from abundant 2D medical images can improve 3D medical VQA when both input types are aligned through a common reasoning interface. W…
Fleming-R1: Toward Expert-Level Medical Reasoning via Reinforcement Learning
Chi Liu, Derek Li, Yan Shu +4
While large language models show promise in medical applications, achieving expert-level clinical reasoning remains challenging due to the need for both accurate answers and transp…
Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling
Derek Li, Jiaming Zhou, Leo Maxime Brunswic +8
The pursuit of general-purpose artificial intelligence depends on large language models (LLMs) that can handle both structured reasoning and open-ended generation. We present Omni-…
Hybrid Open-Ended Tri-Evolution Makes Better Deep Researcher
Hongming Piao, Chi Liu, Mengzhuo Chen +5
Deep research and agent evolution serve as de-facto tasks for AI agents in real-world applications toward artificial general intelligence. The former enables autonomous retrieval a…
Best-of-Evidence: Best-of-N Selection under Partial Verification
Cenwei Zhang, Teng Fang, Yuxia Wang +3
BoN improves model outputs by sampling several candidates and selecting one with a proxy score, but it assumes that complete candidates can be evaluated reliably. Many vision-langu…
Talaria: Session-Aware Serverless Serving of Hundred-Billion-Parameter LLMs
Utopia Meng, Unicornt Zhao, Derek Li +2
Serverless multi-model LLM systems multiplex popularity-skewed model catalogs over shared GPU pools, yet typically schedule each request independently. Tool-using agents break this…
Hierarchical Chain-of-Thought: Enhancing LLM Reasoning Performance and Efficiency
Xingshuai Huang, Derek Li, Bahareh Nikpour +1
Chain-of-Thought (CoT) prompting has significantly improved the reasoning capabilities of large language models (LLMs). However, conventional CoT often relies on unstructured, flat…
Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
Mohammad Ali Alomrani, Yingxue Zhang, Derek Li +14
Large language models (LLMs) have rapidly progressed into general-purpose agents capable of solving a broad spectrum of tasks. However, current models remain inefficient at reasoni…