Search papers, labs, and topics across Lattice.
7
0
9
1
Tool-using agents can be tricked into leaking sensitive data even when each individual tool use seems safe – ChainCaps stops this "permission laundering" with a simple, effective runtime check.
Robots can now understand your gestures, not just your words, leading to more intuitive and efficient human-robot collaboration in cluttered environments.
VLN agents can now navigate more effectively by reasoning about their own state and task progress, closing the gap between end-to-end VLMs and explicit scene mapping.
Today's best vision-language models are surprisingly bad at reading scientific figures, failing to match expert-level reasoning on a new benchmark of experimental images.
Extracting and clustering independent components from EEG data reveals stable neural signatures of emergency braking, enabling more accurate braking intensity prediction.
Generating realistic human-object interaction videos from text, images, audio, *and* pose is now possible, opening the door to automated content creation workflows.
Agentic AI's ability to autonomously retrieve information and invoke tools opens a Pandora's Box of runtime vulnerabilities, including self-propagating "Viral Agent Loops" that bypass traditional code-level exploits.
