Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915211456282624 |
|---|---|
| author | Ng, Wen Zheng Terence Chen, Jianda Xu, Yuan Zhang, Tianwei |
| author_facet | Ng, Wen Zheng Terence Chen, Jianda Xu, Yuan Zhang, Tianwei |
| contents | This work addresses the challenge of personalizing trajectories generated in automated decision-making systems by introducing a resource-efficient approach that enables rapid adaptation to individual users' preferences. Our method leverages a pretrained conditional diffusion model with Preference Latent Embeddings (PLE), trained on a large, reward-free offline dataset. The PLE serves as a compact representation for capturing specific user preferences. By adapting the pretrained model using our proposed preference inversion method, which directly optimizes the learnable PLE, we achieve superior alignment with human preferences compared to existing solutions like Reinforcement Learning from Human Feedback (RLHF) and Low-Rank Adaptation (LoRA). To better reflect practical applications, we create a benchmark experiment using real human preferences on diverse, high-reward trajectories. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_18347 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners Ng, Wen Zheng Terence Chen, Jianda Xu, Yuan Zhang, Tianwei Machine Learning Artificial Intelligence Robotics This work addresses the challenge of personalizing trajectories generated in automated decision-making systems by introducing a resource-efficient approach that enables rapid adaptation to individual users' preferences. Our method leverages a pretrained conditional diffusion model with Preference Latent Embeddings (PLE), trained on a large, reward-free offline dataset. The PLE serves as a compact representation for capturing specific user preferences. By adapting the pretrained model using our proposed preference inversion method, which directly optimizes the learnable PLE, we achieve superior alignment with human preferences compared to existing solutions like Reinforcement Learning from Human Feedback (RLHF) and Low-Rank Adaptation (LoRA). To better reflect practical applications, we create a benchmark experiment using real human preferences on diverse, high-reward trajectories. |
| title | Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners |
| topic | Machine Learning Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2503.18347 |