SKILL2STEER: One-Pass Hypernetwork for Mapping Natural-Language Skills to Low-Rank Post-Block Activation Shifts
Organizations: Hong Kong University of Science and Technology · Hong Kong University of Science and Technology(Guangzhou)
Abstract
Compiling a natural-language agent skill into model parameters removes the skill text from the prompt and leaves behind a small artifact that can be distributed and reused. Existing methods differ in where they inject the compiled quantity, and all of them assume that the artifact carries the content of the skill. We test that assumption. Through three controls on a retrieval agent, we find that a compiled skill shift carries a domain-general behavioural prior, not the content of the skill. A wrong skill document performs on par with the correct one, a single averaged bundle matches per-skill routing under a two-sided equivalence test, and a same-norm random direction collapses to the no-injection baseline, which shows that the trained direction stays load-bearing. We then introduce SKILL2STEER, a one-pass hypernetwork that reads a skill document once and maps it to low-rank activation shifts added at the output of every decoder block. Each skill compiles to 2.36M parameters, 9.25 times fewer than a weight-space adapter, while the base weights stay frozen and the generator runs only at compile time. The shift can be scheduled across retrieval steps and deployed as one routing-free bundle. On SearchQA, a seven-dataset benchmark, SKILL2STEER reaches 37.9 exact match, 9.8 points above the frozen model and 2.3 points above a weight-space method that ships 9.25 times more parameters per skill.
Figures & tables
| Exact match per dataset (%) | Summary | ||||||
| Method | NQ | TriviaQA | PopQA | Avg | Avg | Params / skill | Extra input |
| Prompting, retrieval | |||||||
| Vanilla (base) | 25.2 | 50.6 | 35.2 | 37.0 | ref. | — | ✗ |
| CoT | 19.2 | 50.8 | 20.0 | 30.0 | — | ✓ | |
| Few-Shot | 34.6 | 57.6 | 39.4 | 43.9 | — | ✓ | |
| R1-Instruct | 27.0 | 55.0 | 31.6 | 37.9 | — | ✓ | |
| Exact match per dataset (%) | Summary | |||||||
| Method | HotpotQA | 2Wiki | MuSiQue | Bamboogle | Avg | Avg | Params / skill | Extra input |
| Prompting, retrieval | ||||||||
| Vanilla (base) | 26.2 | 26.8 | 4.2 | 28.8 | 21.5 | ref. | — | ✗ |
| CoT | 22.4 | 26.2 | 5.0 | 37.6 | 22.8 | — | ✓ | |
| Few-Shot | 30.0 | 25.8 | 6.0 | 17.6 | 19.9 | — | ✓ | |
| R1-Instruct | 26.6 | 33.6 | 5.8 | 34.4 | 25.1 | — | ✓ | |
| Method | 1-hop | Multi | All-7 | Params |
|---|---|---|---|---|
| Vanilla (base) | 37.0 | 21.5 | 28.1 | – |
| LatentSkill | 44.9 | 27.0 | 35.6 | 21.8 M |
| Ours (static / static) | 46.6 | 25.7 | 35.7 | 2.36 M |
| Ours (static / sched1) | 46.6 | 30.0 | 37.9 | 2.36 M |
| Ours (6000 steps) | 43.7 | 30.0 | 36.6 | 2.36 M |
| Ours (universal) | 47.1 | 30.2 | 38.3 | 2.36 M |
| Setting | EM | Steps | Inv.% | Chars |
| Single-hop (1500) | ||||
| No injection | 38.5 | 1.38 | 3.33 | 2238 |
| Skill2Steer | 46.6 | 2.15 | 0.03 | 45 |
| Skill2Steer (sched1) | 45.8 | 2.16 | 3.31 | 1411 |
| Multi-hop (1625) | ||||
| No injection | 25.2 | 1.65 | 5.77 | 3287 |
Appendix figures & tables1 asset
Supplementary material from the paper’s appendix.
Appendix
| Setting | EM | Steps | Inv.% | Chars |
|---|---|---|---|---|
| Ours (6000 steps) | 43.7 | 1.93 | 0.00 | 3132 |
| Ours (gen-only, multi) | 14.2 | 1.01 | – | 5108 |
LLM Agent Skill Learning
Large language model agent skill learning studies how agents acquire, represent, reuse, and refine procedures for multi-step tasks.
See all 424 papers