Search papers, labs, and topics across Lattice.
4
0
7
Static sample selection in reinforcement fine-tuning is a recipe for suboptimal updates鈥擠IEM adapts dynamically, leading to superior performance on reasoning tasks.
GIVE achieves reliable geometric edits in video editing by unifying object-state specifications, overcoming traditional challenges in generative workflows.
Forget agonizing over checkpointing and restarts: LiveR slashes LLM training downtime from minutes to seconds by hot-swapping model state between parallel training worlds.
Hopper's idle NVLink Copy Engine can be turned into a nearly free communication channel for MoE load balancing, slashing token stragglers by up to 70% without impacting existing parallelism.
