Search papers, labs, and topics across Lattice.
4
0
9
Forget specialized architectures: StepAudio 2.5 proves a single audio-language foundation, shaped by RLHF, can dominate ASR, TTS, and real-time dialogue simultaneously.
RL fine-tuning unlocks a 6x performance gain for in-place trajectory editing in autonomous driving, demonstrating the power of aligning diffusion planners with reinforcement learning.
LLMs can now tackle complex table QA with 20%+ accuracy gains, thanks to a multi-agent framework that decomposes queries and orchestrates reasoning between specialized database and knowledge graph agents.
Achieve state-of-the-art video polyp segmentation by adaptively selecting informative reference frames and aggregating multi-scale historical features with causal attention.
