Gen Li
9 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
C-MIG is a novel retrieval-augmented generation framework that uses multi-view information gain to improve clinical diagnosis reasoning by providing richer, more nuanced reward signals than existing methods.
The paper proposes EAPO, an entropy-driven adaptive weighting method that dynamically adjusts the influence of positive samples during policy optimization to improve both response diversity and stability in open-ended QA.
C-MIG is a novel retrieval-augmented generation framework that uses multi-view information gain to improve clinical diagnosis reasoning by providing richer, more nuanced reward signals than existing methods.
The paper proposes EAPO, an entropy-driven adaptive weighting method that dynamically adjusts the influence of positive samples during policy optimization to improve both response diversity and stability in open-ended QA.
OccamToken introduces a training-free, adaptive token pruning framework that replaces fixed token budgets with relative evidence testing against a register-based reference, significantly improving VLM efficiency while maintaining high accuracy.
OccamToken introduces a training-free, adaptive token pruning framework that replaces fixed token budgets with relative evidence testing against a register-based reference, significantly improving VLM efficiency while maintaining high accuracy.
This paper proposes GIVE, an approach that enhances VLA models with human gesture understanding to improve Human-Robot Interaction.
The paper proves that diffusion models adapt to low-dimensional structure robustly across various update coefficient choices.
The paper presents GenWorld, an urban simulation infrastructure for LLM-agent studies, combining a synthetic city, agent-environment interface, and offline compilation of LLM-derived signals.
Papers
GenWorld: Empirically Grounded Urban Simulation Infrastructure for Scalable LLM-Agent Studies
Gen Li, Jieyuan Lan, Pengcheng Xu, Zongyuan Wu +2 more
The paper presents GenWorld, an urban simulation infrastructure for LLM-agent studies, combining a synthetic city, agent-environment interface, and offline compilation of LLM-derived signals.