Track new results across pure and applied mathematics.
Search papers, labs, and topics across Lattice.
3
0
4
26
Autonomous agents can now discover their own evaluation metrics, bypassing the costly bottleneck of real-world engineering benchmarks.
Static environments can be transformed on-the-fly to better suit agent learning, resulting in up to a 9.0-point performance boost with fewer execution steps.
Autonomous research agents are surprisingly unreliable, with existing systems hallucinating references 21% of the time and failing to align methods with code as often as 80%, but a new "Chain-of-Evidence" approach can fix this.