Search papers, labs, and topics across Lattice.
4
0
9
0
Clinicians using HeartAgent, a cardiology-specific agent system, improved diagnostic accuracy by 26.9% and explanatory quality by 22.7% compared to unaided experts.
Training embodied intelligence models just got 40x faster thanks to a thousand-GPU cloud platform and a suite of optimizations spanning data pipelines, model architecture, and infrastructure.
LLMs in collaborative coding often stumble on interaction subtleties, leading to a new class of problems called "Interaction Smells" that can now be systematically identified and mitigated.
HybridRAG-Bench reveals that existing benchmarks overestimate the reasoning abilities of retrieval-augmented LLMs due to contamination, offering a more realistic evaluation using up-to-date scientific knowledge.
