Lijun Wu

D DianShi-RxnDB is presented, a large-scale, fine-grained organic reaction data platform built via a fully automated extraction and normalization pipeline integrating patent text, images, and reaction schemes integrating patent text, images, and reaction schemes.

Achieving 93.02% accuracy in molecular structure recognition, MinerU.Chem outperforms existing systems, unlocking new possibilities for AI-driven chemistry research.

Agents trained in this framework achieve unprecedented improvements in reasoning depth and citation accuracy, transforming the landscape of automated peer review processes.

A single multimodal model outperforms specialized approaches across 80 biological tasks, redefining the landscape of biological AI applications.

Unlock long-context reasoning in LLMs by turning agent trajectories into gold-standard QA pairs, outperforming models 8x larger on challenging reasoning tasks.

LLM datasets aren't independent islands: tracing their lineage reveals hidden redundancy, benchmark contamination, and opportunities for more diverse training data.

An 8B model can now generate scientific graphics code that rivals or surpasses the output of much larger proprietary models, thanks to a new dataset, benchmark, and reinforcement learning approach.