Search papers, labs, and topics across Lattice.
7
0
9
2
MLLMs exhibit alarming Stochastic Collapse, failing to maintain randomness even under explicit random instructions, which could undermine their utility in diverse applications.
AdaCodec cuts time-to-first-token from 9.26 seconds to just 1.62 seconds while outperforming traditional per-frame RGB encoding methods.
Frontier models can't build playable games in one shot, but a closed-loop system using GUI agents to playtest and refine code achieves a 66.8% success rate, proving that game generation needs to be a conversation, not a translation.
By co-evolving experts through bidirectional policy distillation, CoPD achieves all-in-one integration of text, image, and video reasoning, outperforming domain-specific experts and suggesting a new training paradigm.
Forget external teachers – the best way to boost your RL model's performance is to learn from its future self.
EasyVideoR1 achieves a 1.47 times throughput improvement in video understanding tasks by eliminating redundant video decoding and leveraging a comprehensive task-aware reward system.
Self-distillation in LLMs can leak information and destabilize training, but combining it with verifiable rewards yields a sweet spot for improved convergence and stability.
