Search papers, labs, and topics across Lattice.
7
127
6
5
MMRAG gets a human-like reasoning upgrade: CogniVerse uses cognitive reflection and information geometry to filter noise, align modalities, and generate coherent responses, outperforming existing systems.
LVLMs are surprisingly susceptible to universal, black-box adversarial attacks that synergistically combine imperceptible image perturbations with subtle text prompts.
LLM defenses don't have to sacrifice performance: APD disentangles adversarial prompts to slash harmful outputs by 85% while maintaining model utility.
VLMs can be significantly improved by reasoning over diverse, generated text inputs, rather than relying on restrictive, predefined templates.
VLMs can now handle real-world sensor failures and data privacy constraints without catastrophic performance drops, thanks to a new plug-and-play module for incomplete multi-modal inputs.
Current video moment retrieval systems fail catastrophically when given irrelevant queries, but this work introduces a method to detect and reject such queries, preventing potentially dangerous false retrievals.
Stop wasting compute on irrelevant video clips: SpotVMR trims videos to only the query-relevant moments, boosting retrieval performance while slashing computational cost.
