Loading activity…
@lianapatel_
CS PhD @Stanford, building systems for llm data processing, including LOTUS (http://github.com/lotus-data/lotus), ACORN and DeepScholar (http://deep-scholar.vercel.app)
AI
The authors report faster time-to-first-token and lower peak memory across diverse tasks, without accuracy degradation.
AI
A study author reports an advantage of up to 37 points, with 4× fewer turns—and asks what systems researchers still need to solve as models improve.
AI
Stanford PhD researcher building LLM data systems will co-organize NOVAS workshop at VLDB.
