Search papers, labs, and topics across Lattice.
7
0
10
3
Inaudible low-frequency signals can cripple LALM performance by up to 67%, revealing a hidden vulnerability in audio processing systems.
Adversarial techniques traditionally seen as threats are now being repurposed by content owners to proactively safeguard their visual assets from misuse.
Projector fine-tuning, commonly used for aligning MLLMs, unexpectedly introduces backdoor vulnerabilities with activation mechanisms distinct from those in text-only LLMs.
GaLa's hypergraph representation reveals hidden semantic relationships in multimodal data, leading to a dramatic boost in procedural planning accuracy.
Over 20 teams vied to decode human attention in video, revealing new insights into saliency prediction techniques.
LALMs are shockingly vulnerable to inaudible audio prompts that can make them execute unauthorized actions, even on commercial systems like Mistral AI and Microsoft Azure.
Audio backdoor attacks leave a tell: triggers are surprisingly stable to destructive noise but fragile to meaning-preserving changes.
