Guided by a learned value critic, 1-step lookahead and trajectory selection lift an open-weights SWE agent to 40.8% on SWE-bench Verified, roughly doubling its success rate.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Guided Search Strategies in Non-Serializable Environments with Applications to Software Engineering Agents
Guided by a learned value critic, 1-step lookahead and trajectory selection lift an open-weights SWE agent to 40.8% on SWE-bench Verified, roughly doubling its success rate.