arXiv:2605.09134v3 Announce Type: replace
Abstract: Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs. We present BoostAPR, a three-stage fra…
arXiv:2605.09134v2 Announce Type: new
Abstract: Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs. We present BoostAPR, a three-stage framewo…
arXiv:2605.09134v1 Announce Type: new
Abstract: Reinforcement learning for program repair is hindered by sparse execution feedback and coarse sequence-level rewards that obscure which edits actually fix bugs. We present BoostAPR, a three-stage framewo…