8 papers
BuddyMoE: Exploiting Expert Redundancy to Accelerate Memory-Constrained Mixture-of-Experts Inference
Yun Wang, Lingyun Yang, Senhao Yu +5
Mixture-of-Experts (MoE) architectures scale language models by activating only a subset of specialized expert networks for each input token, thereby reducing the number of floatin…
Generative Universal Verifier as Multimodal Meta-Reasoner
Xinchen Zhang, Xiaoying Zhang, Youbin Wu +5
We introduce Generative Universal Verifier, a novel concept and plugin designed for next-generation multimodal reasoning in vision-language models and unified multimodal models, pr…
Demystifying Reinforcement Learning in Agentic Reasoning
Zhaochen Yu, Ling Yang, Jiaru Zou +2
Recently, the emergence of agentic RL has showcased that RL could also effectively improve the agentic reasoning ability of LLMs, yet the key design principles and optimal practice…
SD-VLM: Spatial Measuring and Understanding with Depth-Encoded Vision-Language Models
Pingyi Chen, Yujing Lou, Shen Cao +6
While vision language models (VLMs) excel in 2D semantic visual understanding, their ability to quantitatively reason about 3D spatial relationships remains under-explored, due to…
Winning Gold at IMO 2025 with a Model-Agnostic Verification-and-Refinement Pipeline
Yichen Huang, Lin F. Yang
The International Mathematical Olympiad (IMO) is widely regarded as the world championship of high-school mathematics. IMO problems are renowned for their difficulty and novelty, d…
Do LLMs Know When to Flip a Coin? Strategic Randomization through Reasoning and Experience
Lingyu Yang
Strategic randomization is a key principle in game theory, yet it remains underexplored in large language models (LLMs). Prior work often conflates the cognitive decision to random…