UO-FIE: Combining Exact-Label Supervision with Graded Utility for Factivity Inference
arXiv:2609.28605v1 Announce Type: new Abstract: The Factivity Inference Evaluation 2026 (FIE2026) classifies Chinese context-hypothesis pairs into nine ordered factivity intervals. Its evaluation metric rewards both exact predictions and proximity to the correct interval, while 64.1% of the 566 tra…
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-09-25 04:00 · arXiv cs.LG
UO-FIE: Combining Exact-Label Supervision with Graded Utility for Factivity Inference