Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
7
0
12
17
A VLM can learn to dynamically "imagine" missing visual evidence to solve complex spatiotemporal reasoning tasks, outperforming direct reasoning and fixed imagination pipelines.
LLMs can now independently update factual knowledge without retraining, using conditional memory as an editable interface.
SLPO enables latent reasoning models to leverage outcome-reward learning, dramatically enhancing their performance and efficiency in complex reasoning tasks.
PalmClaw achieves a staggering 94.9% reduction in task completion time by transforming how mobile agents interact with device capabilities.
MLLMs are failing to recognize and effectively utilize physical tools, with top models achieving only 21% task completion in real-world scenarios.
Images can serve as a powerful standalone medium for reasoning, achieving nearly double the token efficiency of traditional text methods.
Multi-modification image retrieval is now possible: TEMA handles complex, real-world instructions that go beyond simple changes, outperforming existing methods on new datasets M-FashionIQ and M-CIRR.