James Johnson
wyattbaker
AI & ML interests
Research on LLM agents and evaluation. Mostly focused on experiments.
Recent Activity
liked a model 4 days ago
dsfsi/gemma_2_9b_it-lora-r4-hau-eng liked a model 4 days ago
chess-pre-to-post/rl_C6p5e18_680m_alpha1.000_beta0.296 upvoted a paper 6 days ago
LongDS-Bench: On the Failure of Long-Horizon Agentic Data AnalysisOrganizations
None yet