events
Oct 22 - Advances in AI at Virginia Tech
Tre interventi di ricerca di Virginia Tech sulla comunicazione multiagente tra LLM, sul grounding visivo a grana fine nei piccoli modelli multimodali e sulla generazione controllabile di immagini e…

In inglese
- Starts
- 2026-10-22 16:00 UTC
- Organizer
- Stuttgart AI, Machine Learning and Computer Vision Meetup
- Location
- Online event
Three research talks from Virginia Tech PhD students, hosted online by the Stuttgart AI, Machine Learning and Computer Vision Meetup. The first covers communication in multi-agent LLM systems, including CONSENSAGENT, which targets sycophancy to improve consensus, and work on diagnosing failure modes in multi-agent debate. The second presents a coarse-to-fine, reinforcement-learning approach that lets a compact 4B-parameter multimodal model ground small object parts. The third examines representations inside image and video generative models as a route to precise editing, composition and temporal control.
Event website