Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
3
0
6
Achieving an AUROC of 0.878 on the CHB-MIT benchmark, this multimodal EEG model redefines the landscape of seizure detection by enabling robust performance across diverse datasets and conditions.
Vision-language models struggle with safety-critical reasoning, but AUTOPILOT-VQA reveals their limitations in understanding real-world driving incidents.
Finally, a way to understand *how* transformers make decisions, not just *which* tokens are important, by tracing relevance flow through layers with context-aware gradients.