Flow to Learn: Flow Matching on Neural Network Parameters
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_ | 1866910914918219776 |
|---|---|
| author | Saragih, Daniel Cao, Deyu Balaji, Tejas Santhosh, Ashwin |
| author_facet | Saragih, Daniel Cao, Deyu Balaji, Tejas Santhosh, Ashwin |
| contents | Foundational language models show a remarkable ability to learn new concepts during inference via context data. However, similar work for images lag behind. To address this challenge, we introduce FLoWN, a flow matching model that learns to generate neural network parameters for different tasks. Our approach models the flow on latent space, while conditioning the process on context data. Experiments verify that FLoWN attains various desiderata for a meta-learning model. In addition, it matches or exceeds baselines on in-distribution tasks, provides better initializations for classifier training, and is performant on out-of-distribution few-shot tasks while having a fine-tuning mechanism to improve performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_19371 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Flow to Learn: Flow Matching on Neural Network Parameters Saragih, Daniel Cao, Deyu Balaji, Tejas Santhosh, Ashwin Machine Learning Artificial Intelligence Foundational language models show a remarkable ability to learn new concepts during inference via context data. However, similar work for images lag behind. To address this challenge, we introduce FLoWN, a flow matching model that learns to generate neural network parameters for different tasks. Our approach models the flow on latent space, while conditioning the process on context data. Experiments verify that FLoWN attains various desiderata for a meta-learning model. In addition, it matches or exceeds baselines on in-distribution tasks, provides better initializations for classifier training, and is performant on out-of-distribution few-shot tasks while having a fine-tuning mechanism to improve performance. |
| title | Flow to Learn: Flow Matching on Neural Network Parameters |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2503.19371 |