Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.20760 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917228368101376 |
|---|---|
| author | Shashidhar, Sarvesh Mishra, Abhishek Kotecha, Madhav |
| author_facet | Shashidhar, Sarvesh Mishra, Abhishek Kotecha, Madhav |
| contents | Re-inforcement learning from human feedback (RLHF) has been effective in the task of AI alignment. However, one of the key assumptions of RLHF is that the annotators (referred to as workers from here on out) have a homogeneous response space. This assumption is not true in most practical settings and there have been studies done in the past to challenge this notion. This work has been inspired by such studies and explores one of the ways to deal with heterogeneity in worker preferences - by clustering workers with similar preferences and personalising reward models for each cluster. This work provides an algorithm that encourages simultaneous learning of reward models and worker embeddings. This algorithm is then empirically tested against the Reddit TL;DR dataset with unique worker IDs. We have shown that clustering users into different groups based on their preferences and created personalised reward models improves win-rate of the said models. Along with results and visualisations, this work aims to act as a stepping stone to more complicated models and gives a list of possible future extensions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20760 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Exploring Re-inforcement Learning via Human Feedback under User Heterogeneity Shashidhar, Sarvesh Mishra, Abhishek Kotecha, Madhav Human-Computer Interaction Re-inforcement learning from human feedback (RLHF) has been effective in the task of AI alignment. However, one of the key assumptions of RLHF is that the annotators (referred to as workers from here on out) have a homogeneous response space. This assumption is not true in most practical settings and there have been studies done in the past to challenge this notion. This work has been inspired by such studies and explores one of the ways to deal with heterogeneity in worker preferences - by clustering workers with similar preferences and personalising reward models for each cluster. This work provides an algorithm that encourages simultaneous learning of reward models and worker embeddings. This algorithm is then empirically tested against the Reddit TL;DR dataset with unique worker IDs. We have shown that clustering users into different groups based on their preferences and created personalised reward models improves win-rate of the said models. Along with results and visualisations, this work aims to act as a stepping stone to more complicated models and gives a list of possible future extensions. |
| title | Exploring Re-inforcement Learning via Human Feedback under User Heterogeneity |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2601.20760 |