Search papers, labs, and topics across Lattice.
3
0
5
0
Evolving video agents can boost long-form video understanding accuracy by over 13 percentage points while dramatically reducing processing costs.
Diffusion models get a 2.88x speed boost in low-step regimes thanks to a new Pad茅 approximation technique that predicts feature evolution more accurately than Taylor-based methods.
Even state-of-the-art text-to-image models like Qwen-Image can be significantly improved in structural fidelity and semantic alignment of rendered text using a novel RL strategy that rewards structural anomaly quantification.
