Papers › End-to-End Temporal Relation Extraction in the Clinical Domain
End-to-End Temporal Relation Extraction in the Clinical Domain
José Javier Saiz, Begoña Altuna
Temporal relation extraction is an important task in the clinical domain, as it allows a better understanding of the temporal context of clinical events. In this paper, we present an end-to end temporal relation extraction system for the clinical domain, using the i2b2 2012 Temporal Relation challenge as a benchmark. In our proposal, we fine-tune REBEL —a sequence-to-sequence model for general relation extraction — with temporal annotations and discharge summaries. Our proposal is then able to simultaneously extract relevant clinical entities, time expressions and the temporal relations between them. Our results demonstrate the efectiveness of this approach, achieving reasonable performance on the End-To-End track of the i2b2 2012 Challenge.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
1 archive task tag without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Joint Entity and Relation Extraction | 2012 i2b2 Temporal Relations | Finetuned REBEL | Macro F1 | 0.58 | #1 of 1 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections