Jing Yang

LLMs can now autonomously retrieve relevant memories from a database using specialized tools, significantly improving performance on long-term conversational question answering.

Federated differentially private data synthesis can now achieve utility comparable to centralized approaches, even with heterogeneous data distributions, thanks to a novel framework that smartly handles noise and redundancy.

Slash RAG latency by an order of magnitude using a tiny, LoRA-adapted SLM that routes queries, achieving GPT-4o-mini level accuracy at a fraction of the cost.

LLMs can now reason effectively about complex agricultural scenarios by iteratively writing and executing code within a specialized environment, outperforming traditional text-based approaches.

Forget brittle hyperparameter tuning: a game-theoretic approach adaptively boosts performance on rare labels in extreme multi-label classification by rewarding curiosity about tail labels.