Search papers, labs, and topics across Lattice.
Affiliation:
3
0
6
Direct, parallel, and sequential generation methods reveal distinct strengths and weaknesses in multi-subject video synthesis, with implications for future I2V system designs.
MLLMs can significantly reduce hallucinations and improve reasoning by shifting focus from linguistic shortcuts to causal visual grounding through VIGIL.
Long-term memory doesn't just expand LLMs' knowledge, it also dramatically expands their attack surface: a new black-box attack, ER-MIA, achieves high success rates injecting adversarial memories into LLMs via similarity-based retrieval.
