Search papers, labs, and topics across Lattice.
3
901
6
5
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.
A 4B-parameter model, InternVL-U, outperforms 14B-parameter models in multimodal generation and editing, proving that size isn't everything.
Open-source multimodal models just leveled up: InternVL3 rivals closed-source titans like GPT-4o by pre-training vision and language together from the start.
