Search papers, labs, and topics across Lattice.
4
0
8
19
Subtle timing in subtitle presentation can amplify video jailbreak effectiveness against LVLMs, achieving unprecedented attack success rates.
Contextual information can dramatically amplify the success of semantic-shift jailbreaks, with a new framework achieving a 74.6% attack success rate.
Positional Blind Spots in VLA models can lead to failure rates as high as 58%, but targeted fine-tuning can dramatically reduce these failures by over 40%.
LLMs' impressive code generation skills crumble when faced with the messy reality of ambiguous requirements, highlighting a critical gap in their ability to handle real-world software development scenarios.
