Search papers, labs, and topics across Lattice.
4
0
6
5
This work shows that the frozen agent LLM already carries the routing signal in its own forward passes, and that two linear maps suffice to read it out with no skill text in the context, and shows that Gavel (Glance And Verdict from a frozen LLM) reads it in two steps.
Skill selection in LLMs can be optimized to achieve a 0.73 task success rate while using 28% fewer tokens than existing methods.
Reward function design just got a major upgrade鈥擬LREF achieves 25.2% better performance by reusing and evolving reward components across iterations.
Unleashing an LLM's inner creativity or laser-sharp logic is now as simple as turning a knob, thanks to a new distribution-matching method that avoids heuristic rewards.
