Search papers, labs, and topics across Lattice.
4
0
5
3
Current MLLMs falter in following complex video instructions, revealing a significant oversight in their evaluation metrics.
FIRM-Video reveals that a checklist-driven approach can significantly enhance the reliability of text-to-video reward modeling, achieving state-of-the-art performance in evaluation metrics.
Seedance 2.0 leapfrogs existing models by unifying multi-modal inputs (text, image, audio, video) into a single architecture for generating high-quality, longer-duration audio-video content.
Hallucinations in RL-based image editing and generation are tamed with FIRM, a new framework that trains robust reward models on curated datasets to provide more accurate guidance.
