Long-horizon camera-controlled video generation relies on persistent memory to maintain scene consistency. Existing systems follow two strategies to achieve this consistency. Full-history approaches retain all generated observations, causing unbounded storage and retrieval costs. Selective-construction approaches reduc...
Abdul Mohaimen Al Radi, Kun-Yang Li, Yu-Zhang Shang et al.· 0 citations
Language model agents are increasingly deployed to automate everyday digital chores from managing emails and social media to handling banking and bills allowing users to step away from supervision. However, this capability also exposes sensitive information to phishing. Safe execution requires distinguishing malicious...
Rafid Ahmed, Joseph Fioresi, Mubarak Shah et al.· 0 citations
CineMR is presented, a tool-augmented VLM that invokes cardiac image-analysis tools and integrates their outputs into interleaved reasoning for quantitative CMR assessment and a multi-cohort visual question answering benchmark covering quantitative metric extraction, multiclass diagnosis, and differential diagnosis.
Kun-Yang Li, H. Nguyen, Joshua Lowe et al.· 0 citations
PrivMeSA is introduced, a privacy-aware self-evolving multi-agent system that learns to control disclosure and retains remote expertise for local reuse, and improves mean task accuracy over delegation by up to 15.8 percentage points.
Dan-Nong Wang, Yu-Ran Zhang, Bian-Yin Sun et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.