Learning Personalized Agents from Human Feedback

Dimensions overall score 9.0

GitHub Code Pulse

No public code linked for this paper yet.

• Personalized Agents from Human Feedback (PAHF) framework enables continual personalization by learning online from live interaction using explicit per-user memory.method 90%
• PAHF operationalizes a three-step loop: seeking pre-action clarification, grounding actions in retrieved preferences, and integrating post-action feedback.method 90%
• The PAHF framework was evaluated on embodied manipulation and online shopping benchmarks.method 80%
• PAHF learns substantially faster than no-memory and single-channel baselines.result 85%
• PAHF reduces initial personalization error compared to baselines.result 80%
• PAHF enables rapid adaptation to preference shifts.result 80%
• Performance may vary based on user interaction willingness and quality of feedback.limitation 70%
• Scalability and efficiency of real-time adaptation might be challenging to maintain at large scales.limitation 60%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card