Skill-α uses reinforcement learning with a rollback reward to progressively edit SKILL.md-style procedure files, improving downstream success rates over pipeline baselines on CL-Bench, SpreadsheetBench, and tau2-bench.
Semi-off-policy reinforcement learning for vision-language slow-thinking reasoning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Progressive Agent Skill Generation via Reinforcement Learning
Skill-α uses reinforcement learning with a rollback reward to progressively edit SKILL.md-style procedure files, improving downstream success rates over pipeline baselines on CL-Bench, SpreadsheetBench, and tau2-bench.