Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.02108 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911151009300480 |
|---|---|
| author | Touayouch, Brahim Fosse, Loïc Damnati, Géraldine Lecorvé, Gwénolé |
| author_facet | Touayouch, Brahim Fosse, Loïc Damnati, Géraldine Lecorvé, Gwénolé |
| contents | Multi-task learning (MTL) is often achieved by merging datasets before fine-tuning, but the growing availability of fine-tuned models has led to new approaches such as model merging via task arithmetic. A major challenge in this setting is task interference, which worsens as the number of tasks increases. We propose a method that merges models trained on different tasks into a single model, maintaining strong performance across all tasks. Our approach leverages Jensen-Shannon divergence to guide the merging process without requiring additional labelled data, and automatically balances task importance. Unlike existing methods, our approach remains robust as the number of tasks grows and consistently outperforms prior work. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_02108 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DivMerge: A divergence-based model merging method for multi-tasking Touayouch, Brahim Fosse, Loïc Damnati, Géraldine Lecorvé, Gwénolé Machine Learning Multi-task learning (MTL) is often achieved by merging datasets before fine-tuning, but the growing availability of fine-tuned models has led to new approaches such as model merging via task arithmetic. A major challenge in this setting is task interference, which worsens as the number of tasks increases. We propose a method that merges models trained on different tasks into a single model, maintaining strong performance across all tasks. Our approach leverages Jensen-Shannon divergence to guide the merging process without requiring additional labelled data, and automatically balances task importance. Unlike existing methods, our approach remains robust as the number of tasks grows and consistently outperforms prior work. |
| title | DivMerge: A divergence-based model merging method for multi-tasking |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.02108 |