Papers › ITER: Iterative Transformer-based Entity Recognition and Relation Extraction
ITER: Iterative Transformer-based Entity Recognition and Relation Extraction
Moritz Hennen, Florian Babl, Michaela Geierhos
When extracting structured information from text, recognizing entities and extracting relationships are essential. Recent advances in both tasks generate a structured representation of the information in an autoregressive manner, a time-consuming and computationally expensive approach. This naturally raises the question of whether autoregressive methods are necessary in order to achieve comparable results. In this work, we propose ITER, an efficient encoder-based relation extraction model, that performs the task in three parallelizable steps, greatly accelerating a recent language modeling approach: ITER achieves an inference throughput of over 600 samples per second for a large model on a single consumer-grade GPU. Furthermore, we achieve state-of-the-art results on the relation extraction datasets ADE and ACE05, and demonstrate competitive performance for both named entity recognition with GENIA and CoNLL03, and for relation extraction with SciERC and CoNLL04.
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
2 archive task tags without a task page not shown.
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Relation Extraction | ACE 2005 | ITER | Cross Sentence | Yes | #1 of 30 | Archive leaderboard | report |
| Relation Extraction | ACE 2005 | ITER | NER Micro F1 | 91.6 ± 0.12 | #1 of 30 | Archive leaderboard | report |
| Relation Extraction | ACE 2005 | ITER | RE Micro F1 | 75.1 ± 0.49 | #1 of 30 | Archive leaderboard | report |
| Relation Extraction | ACE 2005 | ITER | RE+ Micro F1 | 71.9 ± 0.56 | #1 of 30 | Archive leaderboard | report |
| Relation Extraction | ACE 2005 | ITER | Sentence Encoder | FLAN T5 3B | #1 of 30 | Archive leaderboard | report |
| Relation Extraction | Adverse Drug Events (ADE) Corpus | ITER | NER Macro F1 | 92.63 ± 0.89 | #1 of 15 | Archive leaderboard | report |
| Relation Extraction | Adverse Drug Events (ADE) Corpus | ITER | RE+ Macro F1 | 85.6 ± 1.42 | #1 of 15 | 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