Visual General Intelligence white paper authors to present at Video Model Journal Club
HirokatuKataoka · x · 2026-09-18
The Video Model Journal Club hosts Kohsuke Ide (Univ. of Tsukuba / AIST) on Sep 18, 7:30 PM PT, presenting his co-authored 'Visual General Intelligence: A White Paper' with Hirokatu Kataoka and others.
The talk asks which visual differences should change a model's answer and which shouldn't, drawing on work on learning relations between 3D objects and color-dependent judgments in vision-language models, fine-grained geometry, and how text presentation affects results. He closes with a research challenge: can a system learn to obtain useful evidence through observation and interaction when current observations are insufficient, and can those ways of finding out transfer to unfamiliar situations?
More from Research
- Virtual Biotech runs 37,075 parallel AI agents to scale drug discovery like a research org — bravo_abad · 2026-09-18
- AMP reframes robot manipulation as pixel classification to dodge action-space explosion — jiqizhixin · 2026-09-18
- PWNN Paper: Neural Networks as Remote Fault Injectors to Steal AES Keys From Cloud FPGAs — chaumian · 2026-09-18
- Camera-Only Autonomous Drive Through the Pyrenees: Lessons From a Small PhD Team — abursuc · 2026-09-18
- ACE-Data-0 hits 120K downloads in first month: 17M-frame multimodal embodied AI dataset — liuziwei7 · 2026-09-18
- Fei-Fei Li: 540 million years of vision evolution drove the development of intelligence — rohanpaul_ai · 2026-09-18