Search papers, labs, and topics across Lattice.
Affiliation:
4
0
6
Even leading MLLMs struggle with object hallucination, revealing critical integration-stage weaknesses that standard benchmarks fail to capture.
Harmful prompts can slip through safety nets when embedded in benign batches, revealing a critical vulnerability in current AI safety measures.
A unified action embedding space across diverse humanoid robots boosts performance and interpretability, transforming how we approach robot policy learning.
Mapping LLM attack strategies onto a multiplex network reveals interpretable vulnerability clusters and dramatically improves red teaming efficiency.
