WTF
what the fnord
deep dives
glossary
about
feed
2026-03-09
KARL: Knowledge Agents via Reinforcement Learning
Databricks trained an RL-based search agent on GLM 4.5 Air that beats Claude 4.6 and GPT 5.2 on enterprise knowledge retrieval — at a fraction of the cost.
9 min
paper-deep-dives