AI.info

Jinlan Fu

Explore Jinlan Fu on AI.info.

• HERMES: KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding
• FutureOmni: Evaluating Future Forecasting from Omni-Modal Context for Multimodal LLMs
• $\mathcal{V}isi\mathcal{P}runer$: Decoding Discontinuous Cross-Modal Dynamics for Efficient Multimodal LLMs