Search papers, labs, and topics across Lattice.
6
0
10
9
Quantizing rollouts in LLM RL pipelines introduces a training-inference gap that QaRL closes, leading to +5.5 performance on math problems.
Achieve near-lossless 2-bit LLMs with a novel quantization-aware training scheme that progressively reduces precision and intelligently handles outlier channels.
Control both multi-subject identity and multi-granularity motion in video generation with DreamVideo-Omni, a framework that uses latent identity reinforcement learning to avoid identity degradation.
Strategic data curation using a dual-consensus approach beats brute-force training on large noisy datasets for process reward modeling in biological reasoning.
LLMs can now achieve better memory coherence and response fidelity thanks to MemFly's information bottleneck approach to on-the-fly memory optimization.
MLLMs can be made significantly safer in multi-turn dialogues with a new framework that combines cold-start refusal and turn-aware policy optimization, achieving a 10% drop in attack success rate.
