The Router Within: Eliciting Native Skill Routing from a Frozen LLM
Gavel reads skill-routing signals from a frozen LLM's forward passes with two linear maps, beating retrieval pipelines by up to 21.9 points.
Gavel (Glance And Verdict) shows a frozen agent LLM already contains skill-routing signals in its forward passes, read out via two trained linear maps without loading skill text into context. A glance step scores the full library using mid-layer states and per-skill banks built in one forward pass; a verdict step fuses the model's own likelihood and yes/no judgment as a product of experts. Trained once, it transfers zero-shot to three public benchmarks and SkillTraj (372 simulated agent trajectories); on Qwen3-32B it beats progressive disclosure and retrieve-and-rerank pipelines adding 1.2B-16B external parameters by up to 13.4 points (21.9 mid-rollout).
- Frozen LLM's forward passes contain native skill-routing signal
- Only two linear maps trained; no skill text in context
- Outperforms retrieve-and-rerank by up to 21.9 points on Qwen3-32B
- Introduces SkillTraj benchmark of 372 simulated agent trajectories
From the original article
1 min read at arxiv.orgSkills extend an LLM agent beyond its parametric knowledge, and the gain they promise rests on picking the right one. Deployed harnesses route by preloading every skill's metadata into the context, which disperses the agent's attention and caps the library size. Retrieval pipelines move the selection out of the context, but also out of the agent's capability.
By Ruishuo Chen · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.