Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.08070 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911014753140736 |
|---|---|
| author | Qin, Ziheng Xu, Hailun Yew, Wei Chee Jia, Qi Luo, Yang Sarkar, Kanchan Guan, Danhui Wang, Kai You, Yang |
| author_facet | Qin, Ziheng Xu, Hailun Yew, Wei Chee Jia, Qi Luo, Yang Sarkar, Kanchan Guan, Danhui Wang, Kai You, Yang |
| contents | Machine learning relies heavily on data, yet the continuous growth of real-world data poses challenges for efficient dataset construction and training. A fundamental yet unsolved question is: given our current model and data, does a new data (sample/batch) need annotation/learning? Conventional approaches retain all available data, leading to non-optimal data and training efficiency. Active learning aims to reduce data redundancy by selecting a subset of samples to annotate, while it increases pipeline complexity and introduces bias. In this work, we propose Info-Coevolution, a novel framework that efficiently enables models and data to coevolve through online selective annotation with no bias. Leveraging task-specific models (and open-source models), it selectively annotates and integrates online and web data to improve datasets efficiently. For real-world datasets like ImageNet-1K, Info-Coevolution reduces annotation and training costs by 32\% without performance loss. It is able to automatically give the saving ratio without tuning the ratio. It can further reduce the annotation ratio to 50\% with semi-supervised learning. We also explore retrieval-based dataset enhancement using unlabeled open-source data. Code is available at https://github.com/NUS-HPC-AI-Lab/Info-Coevolution/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_08070 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Info-Coevolution: An Efficient Framework for Data Model Coevolution Qin, Ziheng Xu, Hailun Yew, Wei Chee Jia, Qi Luo, Yang Sarkar, Kanchan Guan, Danhui Wang, Kai You, Yang Machine Learning Artificial Intelligence Machine learning relies heavily on data, yet the continuous growth of real-world data poses challenges for efficient dataset construction and training. A fundamental yet unsolved question is: given our current model and data, does a new data (sample/batch) need annotation/learning? Conventional approaches retain all available data, leading to non-optimal data and training efficiency. Active learning aims to reduce data redundancy by selecting a subset of samples to annotate, while it increases pipeline complexity and introduces bias. In this work, we propose Info-Coevolution, a novel framework that efficiently enables models and data to coevolve through online selective annotation with no bias. Leveraging task-specific models (and open-source models), it selectively annotates and integrates online and web data to improve datasets efficiently. For real-world datasets like ImageNet-1K, Info-Coevolution reduces annotation and training costs by 32\% without performance loss. It is able to automatically give the saving ratio without tuning the ratio. It can further reduce the annotation ratio to 50\% with semi-supervised learning. We also explore retrieval-based dataset enhancement using unlabeled open-source data. Code is available at https://github.com/NUS-HPC-AI-Lab/Info-Coevolution/. |
| title | Info-Coevolution: An Efficient Framework for Data Model Coevolution |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2506.08070 |