Chelsea Finn
Chelsea Finn is an Assistant Professor at Stanford University, researching meta-learning and robotics.
Last updated 11 Nov 2026 · Named in 27 answers across 5 engines
Bing AI Mode starts a separate baseline. Copilot remains a separate public engine and keeps its history; changes across the switch are not like-for-like growth.
How AI engines see them
All six engines name Chelsea Finn and agree she is a Stanford professor known for meta-learning, robot learning and reinforcement learning. ChatGPT, Gemini and Google AI Mode credit her with Direct Preference Optimization or MAML. Copilot adds its EXPO-FT system for reinforcement learning finetuning of robot models and SPD for dexterity. Perplexity adds that she leads the IRIS lab. The pages cited most are arxiv.org, ai.stanford.edu and alphaxiv.org.
- ChatGPTNamed in 5 answers
Says Finn is best known for meta-learning, robotics and reinforcement learning, and co-authored Direct Preference Optimization, a simpler alternative to the usual RLHF pipeline.
Known for- meta-learning
- robot learning
- Direct Preference Optimization coauthor
- GeminiNamed in 4 answers
Says Finn is a Stanford professor and Physical Intelligence co-founder who introduced MAML and co-authored the Direct Preference Optimization paper.
Known for- Model-Agnostic Meta-Learning
- Direct Preference Optimization coauthor
- fast model adaptation
- Google AI ModeNamed in 5 answers
Says Finn is a Stanford professor and Physical Intelligence co-founder who co-authored the MAML framework with Pieter Abbeel and Sergey Levine.
Known for- Model-Agnostic Meta-Learning
- fast adaptation with little data
- general-purpose robot learning
- PerplexityNamed in 6 answers
Says Finn is a Stanford professor who leads the IRIS lab, co-founded Physical Intelligence and works on post-training of robot foundation models.
Known for- robot learning
- practical post-training of robot policies
- meta-learning
- Bing AI ModeNot named
No description yet.
- CopilotNamed in 7 answers
Says Finn co-developed EXPO-FT for sample-efficient reinforcement learning finetuning of robot models and introduced simulation pre-training for dexterity.
Known for- EXPO-FT finetuning system
- simulation pre-training for dexterity
- Adaptive Length Penalty
Top Segments
- #11Roboticsof 38Named in 28 answers
Top URLs
- 1Top Robotics & Physical AI Researchers (2026) | Chang Xuchangxu.dev18 citations
- 2The Top 100 Robotics Leaders in the World - The Tech Edvocatethetechedvocate.org6 citations
- 3People | Robot Observatoryrobotobservatory.com5 citations
Prompts where Chelsea is named
Run now opens the answer page for that prompt.
Who are the most influential voices in Post-training & fine-tuning?
Named by 3 engines
Track Chelsea
Follow Chelsea's measured AI visibility and see the questions where they rank.