- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
-Skill trains a single policy with RL to build, execute, update and keep executable tool skills, without supervised skill operations. Training has two stages: a progressive curriculum, followed by leave-one-scenario-out adaptation. The method is evaluated on AppWorld and BFCL-v3.
Critical strengths
- The problem is well motivated, and the two-stage design is coherent.
- The analyses are informative: textual vs. executable skills, skill-management decisions, and execution cost.
- The case study clearly illustrates the skill lifecycle.
Critical weaknesses
- Small absolute gains. The headline relative improvements amount to about 6 tasks on AppWorld and 6 problems on BFCL. SAGE is better in Independent Transfer.
- Unclear attribution. No variant disables skills at test time, and the tight turn budget may favor skills simply because they compress calls.
- Incremental contribution. RL-trained executable skills (SAGE) and RL-based skill evolution (SkillForge, WER, Skill-) already exist, and the closely related methods are not compared experimentally.
Questions for authors
- Where do the Stage I reference trajectories come from?
- How does performance change with skills disabled at test time, or with a larger turn budget?
Thank you for the response. I have read the rebuttal; the new ablations address my attribution concern (W2) and Q1. My assessment of the contribution and effect size is unchanged, so I maintain my rating.