ModsNet: Performance-aware Top-k Model Search using Exemplar Datasets
Summary: ModsNet: a "query-by-example" top-k recommender that, given an exemplar dataset, task and metric, predicts and ranks pretrained models by expected performance using a performance knowledge graph synchronized with a bipartite GNN. Addresses strict cold-start with a cost-bounded probe-and-select strategy that incrementally probes promising models to reduce inference cost and enable prompt, performance-aware model search. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Mengying Wang
- 2. Hanchao Ma
- 3. Sheng Guan
- 4. Yiyang Bian
- 5. Haolai Che
- 6. Abhishek Daundkar
- 7. Alp Sehirlioglu
- 8. Yinghui Wu
Incoming Citations (Sorted by Pagerank)
Showing 1 of 1 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,842 | ML-Asset Management: Curation, Discovery, and Utilization | 2025 | VLDB | 4.1945683e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 0 of 0 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next