WHALE: a simple recipe to jointly optimize an LLM's weights and harness
lateinteraction · x · 2026-09-04
Researcher yoonholeee introduces WHALE, a simple recipe for jointly optimizing an LLM's weights and its harness, with blog and paper released.
- Key observation: harness optimization is sample-efficient but plateaus.
- If you can afford to update the model too, jointly optimizing weights and harness breaks through that ceiling.
Related event: WHALE Paper Proposes Alternating Weight and Harness Optimization(7 posts)→
More from coding & agent
- Agentic SDLC: engineers shift from implementers to agent orchestrators as cycles shrink from weeks to hours — Pavan_Belagatti · 2026-10-21
- OpenAI's dots can now use company tools via Merge for Workforce with employee-bound permissions — shensi · 2026-10-21
- The topology of an LLM agent: where the model, loop, and tools run shapes cost and security — davidcrawshaw · 2026-10-21
- Agents with full-disk and accessibility access are an enterprise security nightmare, argues Runlayer — leveragedupside · 2026-10-21
- Acellera runs an unsupervised virtual lab of AI scientists on a real IL-23 molecular discovery program — gdefabritiis · 2026-10-21
- Pydantic author experiments with JS/TS and React support in the Monty sandbox — samuelcolvin · 2026-10-21