Search papers, labs, and topics across Lattice.
4
0
7
Recursive self-improvement in Macaron-V1 leads to continual learning that adapts to real-world experiences, setting a new standard for open agent models.
LongStraw enables RL post-training with over 2 million tokens on a fixed GPU budget, pushing the boundaries of context length in AI applications.
PEFT can enable the creation of millions of personalized models, each with unique adaptations, leveraging the power of trillion-parameter foundation models.
Personal agents can now dynamically synthesize UIs directly from dialogue context, achieving state-of-the-art results without relying on explicit schemas.
