Search papers, labs, and topics across Lattice.
6
0
7
6
APEx achieves a remarkable 14.7-point improvement over GPT-5.4 by effectively distilling procedural skills for adaptive reasoning in complex question answering.
ZipTok3D reconstructs complex 3D objects with up to 32 times fewer tokens than traditional methods, revolutionizing the efficiency of 3D generation.
SafeAtlas-VL reveals that nuanced, multi-level safety assessments can significantly enhance the performance of multimodal safety models, outperforming existing benchmarks by 4%.
Source-grounded evidence boosts historical entity reconciliation accuracy by up to 94.44 percentage points, revealing the critical role of provenance in resolving identity ambiguities.
Achieving a $5.15\times$ speedup in text-to-3D generation without sacrificing quality could revolutionize real-time 3D content creation.
Text-to-video models can now learn geometrically consistent world dynamics via reinforcement learning, without expensive architectural changes.
