8 papers · 1 filter
Start Classifying: Categorical Critics for LLM Reinforcement Learning
Zhijian Zhou, Long Li, Xuan Zhang +7
Proximal Policy Optimization (PPO) for large language models typically trains its critic by mean-squared-error (MSE) regression on scalar value targets. Although scalar MSE is stat…
Nested Spatio-Temporal Time Series Forecasting
Yinghao Ai, Yukai Zhou, Ruoxi Jiang +8
Spatiotemporal forecasting is critical for real-world applications like traffic management, yet capturing reliable interactions remains challenging under noisy and non-stationary c…
Equivariant Asynchronous Diffusion: An Adaptive Denoising Schedule for Accelerated Molecular Conformation Generation
Junyi An, Chao Qu, Yun-Fei Shi +3
Recent 3D molecular generation methods primarily use asynchronous auto-regressive or synchronous diffusion models. While auto-regressive models build molecules sequentially, they'r…
DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay
Long Li, Zhijian Zhou, Tianyi Wang +7
While Reinforcement Learning (RL) enhances Large Language Model reasoning, on-policy algorithms like GRPO are sample-inefficient as they discard past rollouts. Existing experience…
Unleashing Flow Policies with Distributional Critics
Deshu Chen, Yuchen Liu, Zhijian Zhou +2
Flow-based policies have recently emerged as a powerful tool in offline and offline-to-online reinforcement learning, capable of modeling the complex, multimodal behaviors found in…
The Choice of Divergence: A Neglected Key to Mitigating Diversity Collapse in Reinforcement Learning with Verifiable Reward
Long Li, Zhijian Zhou, Jiaran Hao +9
A central paradox in fine-tuning Large Language Models (LLMs) with Reinforcement Learning with Verifiable Reward (RLVR) is the frequent degradation of multi-attempt performance (Pa…
Guiding Diffusion Models with Reinforcement Learning for Stable Molecule Generation
Zhijian Zhou, Junyi An, Zongkai Liu +5
Generating physically realistic 3D molecular structures remains a core challenge in molecular generative modeling. While diffusion models equipped with equivariant neural networks…
ChemAmp: Amplified Chemistry Tools via Composable Agents
Zhucong Li, Powei Chang, Jin Xiao +6
Although LLM-based agents are proven to master tool orchestration in scientific fields, particularly chemistry, their single-task performance remains limited by underlying tool con…