Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Task Gradient Conflict on Sequential Fine-Tuning Suite (CodeAlpaca, LogiQA, Alpaca) (test)

13Code Conflict Score

EPI

11.7620.1328.536.87Apr 15, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
13119879.6
2026.04
1412119810.8
2026.04
1512109810.8
2026.04
1613119811.4
2026.04
171513111013.2
2026.04
181513121113.8
2026.04
252321191821.2
2026.04
282523212023.4
2026.04
282623222023.8
2026.04
302725232025
2026.04
312826242226.2
2026.04
333027252227.4
2026.04
393633302833.2
2026.04
423836333035.8
2026.04
444037353237.6