Search papers, labs, and topics across Lattice.
9
0
10
12
Transforming historical sequences into a powerful resource, PraMem significantly improves long-horizon behavior prediction beyond existing methods.
LaME achieves 60x faster inference than traditional Chain-of-Thought methods while outperforming some of them in embedding performance.
Achieving lossless processing of 256K contexts, Keye-VL-2.0 transforms how we approach long-video understanding and agentic intelligence.
Surprisingly, the "think before answer" paradigm fails to enhance generative recommendation models, prompting a novel approach that redefines how reasoning is integrated into these systems.
Key contribution not extracted.
MLLMs can't grasp metaphors in videos, revealing a surprising gap in their high-order cognitive abilities compared to humans.
LLMs exhibit a "Utopian bias" when simulating human behavior, converging towards an unrealistic "positive average person" and failing to capture individual differences and long-tail behaviors.
Context-augmented RL lets smaller MLLMs punch *way* above their weight, rivaling much larger models on reasoning tasks while dodging reward hacking.
Unleashing diffusion models' spatial reasoning potential is now possible without expensive joint training, thanks to a clever plug-and-play framework that leverages MLLMs for layout planning.
