Pascale Fung's ICML 2026 keynote: world models, not generative models, for real-world agents
pascalefung · x · 2026-09-16
Pascale Fung (Co-founder of AMI Labs, formerly Senior Director of AI Research at Meta-FAIR) has released the recording of her ICML 2026 invited talk, "Towards AI Agents In the Real World."
Key arguments:
- Today's AI agents are driven by imitation learning with RL on large-scale generative models—strong online but weak in physical-world settings;
- She argues for agents grounded in world modeling: integrating multimodal perception, reasoning-based planning and control, and memory to understand the physical environment, user intentions, and social context;
- Advanced machine intelligence requires modeling both the physical world and the mental world, including latent variables like intent, attention, and context;
- Her world-modeling research grew out of generative models' hallucination limits, and she outlines key challenges toward context-aware, interactive real-world agents.
More from AGI Musings
- Palmer Luckey asks: can you use datacenter RAM for things prohibited by the operator's terms of service? — avaitopiper · 2026-11-02
- On Terence Tao's blog, Frank Calegari calls on mathematicians to adapt after OpenAI's 700-paper release — Hesamation · 2026-11-02
- Polsia's Ben argues AI can autonomously run any human economic activity, including 80% of a company — LinusEkenstam · 2026-11-02
- AI Assistants Remember You, But Nothing Helps You Find What They Told You — Ok_Negotiation_2587 · 2026-11-02
- OpenAI and Anthropic claim AI-discovered proofs disprove the Exponential Time Hypothesis — emmanuelvivier · 2026-11-02
- Researcher: using a $20 AI subscription to fix my phone hints AI may suppress GDP — stanislavfort · 2026-11-02