Search papers, labs, and topics across Lattice.
Affiliation:
4
0
7
Standard intrinsic rewards can fail to drive optimal exploration, even in simple settings, by neglecting the value of counterfactual information.
Closed-loop vision verification is no longer bottlenecked by intractable generative surrogates: compact stochastic world models cut parameter counts by 130脳 while resolving over 80% of previously unverified RGB state spaces.
CryptoProver can independently verify cryptographic libraries in under 12 hours, ensuring the integrity of critical code without altering its execution.
Naively scaling test-time compute is wasteful; strategically allocating it with DIRECT can enhance embodied agent performance while slashing latency by up to 65%.
