News
- 2025-02 Introducing TeleRAG, a lookahead retrieval method for efficient RAG inference! LongLeader, a comprehensive benchmark for long-context models, was accepted by ACL 2025!
- 2024-12 Introducing Explicit Working Memory (EWE) for long-form factual generation! We augment the LLM with a dynamic KV-cache working memory that receives real-time feedback from external resources. Introducing ICONS for multimodal data selection!
- 2024-11 Introducing OpenScholar , an AI assistant proficient at answering research questions with accurate paper citations using online paper resources! Try out our demo!
- 2024-10 Introducing RoRA-VLM, a multimodal retrieval-augmented vision-language model.
- 2024-09 Joined Meta as a visiting researcher!
- 2024-07 Introducing MassiveDS , the first open-source trillion-token datastore and datastore scaling study! We release codes for distributed indexing and API serving make datastore scaling study accessible. [code]
- 2024-06 Started internship at Meta FAIR!
- 2023-10 Gave a talk at AWS on Long-context LLMs! [slides]
- 2023-10 Gave a talk at SAMPL Lab on scaling up retrieval-based language models!
- 2023-10 Introducing LightSeq , a better sequence parallelism solution for long-context LLM training! It is highly optimized for decoders, and it does not entangle with the model architecture, allowing you to scale up your sequence length without an upper bound!
- 2023-09 Started PhD at University of Washington!
- 2023-06 Introducing our latest work on the long-context models (LongChat) and benchmark (LongEval) in this LMSys blog ! Stay tuned!
- 2023-01 Joined AWS as a full-time Applied Scientist working on language model pretraining!
|