Wei Lin

On-policy LLM post-training's generalization advantage stems from its dynamic parameter update direction, a principle transferable to SFT for improved performance and efficiency.

Complex developmental branching and population extrapolation can be resolved with closed-form linear latent operators rather than brittle neural ODEs by framing cell memory through Koopman-Mori-Zwanzig theory.

Up to 47% of reported fine-tuning gains in tool retrieval are an evaluation illusion caused by benchmarks treating functionally equivalent tools as retrieval failures.

CHAP achieves a breakthrough in generative retrieval by aligning dynamic queries with static item representations, enhancing both relevance and inference efficiency.

ATLAS reveals that understanding both immediate request trajectories and long-term user interactions can dramatically enhance the performance of industrial tool-use agents.

By harnessing user behavior trajectories, B2T-Agent dramatically enhances travel plan personalization, outperforming leading models like GPT-4.1.

UniMem achieves a remarkable 4.0 EM point improvement in task execution fidelity by seamlessly integrating episodic and parametric memory without the need for explicit task labels.