Search papers, labs, and topics across Lattice.
Affiliation:
2
0
3
Naive agent-generated test feedback degrades SWE-bench performance by reinforcing shared hallucinations, but decoupling test generation from repair through role-specific RL converts a 3.9-point loss into an 11.4-point gain on open-weights models.
Tool architecture can enhance coding agent performance, with structured interfaces yielding up to 4.7 times more consistency and 41.6% fewer steps in task execution.
