Search papers, labs, and topics across Lattice.
Affiliation:
7
0
8
A single visual world model can now execute complex procedural tasks end-to-end, adapting its generation to the actual outcome of each predicted action.
Frontier generalist systems have effectively mastered high-level visual semantics and spatial reasoning, yet they still stubbornly fail to displace dedicated models on metric geometry, dense temporal tracking, and fine-grained visual fidelity.
Achieving a staggering 79x reduction in decoding latency, Parallel Tube Decoding redefines efficiency in spatio-temporal video grounding.
Multi-view MRI inputs can enhance spatial localization but may compromise temporal reasoning, revealing critical limitations in current foundation models for clinical use.
Expensive end-to-end omni pre-training degrades foundational reasoning and vision, yet simply routing timestamped ASR transcripts into a frozen VLM matches or beats native omni architectures across 56 benchmarks.
A simple training framework boosts pixel-level tampering detection performance in VLMs by over 26%, showcasing the power of balanced sampling and late-injection strategies.
