Track synthesis, materials and catalysis as it is published.
Search papers, labs, and topics across Lattice.
2
0
4
Open-source models can now generate complex, multi-modal computer-use environments that push frontier agent capabilities and enable broad generalization without proprietary frontier models.
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.