Search papers, labs, and topics across Lattice.
4
20
8
5
AgentDoG 1.5 proves you can achieve GPT-5.4-level agent safety with open-source models trained on just 1k samples, slashing deployment overhead by two orders of magnitude.
A 7B model trained on a new dataset of Chinese porcelain outperforms GPT-4 by 12% on expert connoisseurship tasks, demonstrating the power of domain-specific training and tool integration.
Mismatched levels of "mind-reading" between AI agents tank their ability to collaborate, but a simple adaptive strategy can fix it.
LLMs can learn better from human feedback by exploring more creatively, thanks to a simple coin-flip counting method that encourages them to try new things.
