Browse State-of-the-Art › Relation Extraction

Relation Extraction

735 papers with code · 50 benchmarks · 82 datasets archive 2025-07-28

Natural Language Processing

Relation Extraction is the task of predicting attributes and relations for entities in a sentence. For example, given a sentence “Barack Obama was born in Honolulu, Hawaii.”, a relation classifier aims at predicting the relation of “bornInCity”. Relation Extraction is the key component for building relation knowledge graphs, and it is of crucial significance to natural language processing applications such as structured search, sentiment analysis, question answering, and summarization.

Source: Deep Residual Learning for Weakly-Supervised Relation Extraction

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

51 leaderboard tables shown for this task, 50 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 51 until expanded.

Syntology 5 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): DocRED (2 judged, 0 placed), SemEval-2010 Task-8 (1 judged, 0 placed), FUNSD (1 judged, 0 placed), ReDocRED (1 judged, 0 placed), SciERC (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
DocRED (62 rows) DREEAM DREEAM: Guiding Attention with Evidence for Improving... code — Compare
TACRED (40 rows) RAG4RE Retrieval-Augmented Generation-based Relation Extraction code — Compare
SemEval-2010 Task-8 (31 rows) SP Relation Classification as Two-way Span-Prediction — — Compare
ACE 2005 (30 rows) ITER ITER: Iterative Transformer-based Entity Recognition and Relation... code — Compare
CoNLL04 (16 rows) REBEL REBEL: Relation Extraction By End-to-end Language generation code — Compare
Adverse Drug Events (ADE) Corpus (15 rows) ITER ITER: Iterative Transformer-based Entity Recognition and Relation... code — Compare
WebNLG (14 rows) UniRel UniRel: Unified Representation and Interaction for Joint... code — Compare
ChemProt (13 rows) SciBert (Finetune) SciBERT: A Pretrained Language Model for Scientific Text code — Compare
NYT11-HRL (12 rows) RERE Revisiting the Negative Data of Distantly Supervised Relation Extraction code — Compare
ACE 2004 (11 rows) PL-Marker Packed Levitated Marker for Entity and Relation Extraction code Syntology official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 5 where Syntology's instrument failed) · 3 unverified Compare
CDR (10 rows) SAISORE+CR+ET-SciBERT SAIS: Supervising and Augmenting Intermediate Steps for... code — Compare
NYT10-HRL (10 rows) ReRe Revisiting the Negative Data of Distantly Supervised Relation Extraction code — Compare
FUNSD (9 rows) LayoutLMv3 large EM + BBO + RSF A LayoutLMv3-Based Model for Enhanced Relation Extraction in... — — Compare
GDA (9 rows) SAISORE+CR+ET-SciBERT SAIS: Supervising and Augmenting Intermediate Steps for... code — Compare
Re-TACRED (9 rows) EXOBRAIN Improving Sentence-Level Relation Extraction through Curriculum Learning — — Compare
NYT (8 rows) ReLiK-Large ReLiK: Retrieve and LinK, Fast and Accurate Entity Linking and... code — Compare
ReDocRED (8 rows) TTM-RE TTM-RE: Memory-Augmented Document-Level Relation Extraction code Syntology official (archive's flag): 3 ran · 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 3 samples that ran constructed an object rather than computing a result Compare
NYT Corpus (7 rows) KGPOOL KGPool: Dynamic Knowledge Graph Context Selection for Relation Extraction code — Compare
SemEval 2018 Task 10 (6 rows) SVM with GloVe SUNNYNLP at SemEval-2018 Task 10: A Support-Vector-Machine-Based... code — Compare
NYT21 (4 rows) ReRe Revisiting the Negative Data of Distantly Supervised Relation Extraction code — Compare
DDI (3 rows) BioLinkBERT (large) LinkBERT: Pretraining Language Models with Document Links code Syntology official: harvested, nothing ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified Compare
DWIE (3 rows) REXEL REXEL: An End-to-end Model for Document-Level Relation Extraction... code Syntology official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified Compare
GAD (3 rows) BioLinkBERT (large) LinkBERT: Pretraining Language Models with Document Links code Syntology official: harvested, nothing ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified Compare
NYT-single (3 rows) ETL-Span Joint Extraction of Entities and Relations Based on a Novel... code — Compare
NYT29 (3 rows) WDec Effective Modeling of Encoder-Decoder Architecture for Joint... code — Compare
SciERC (3 rows) SciBERT (SciVocab) SciBERT: A Pretrained Language Model for Scientific Text code — Compare
SKE (3 rows) ReRe (exact) Revisiting the Negative Data of Distantly Supervised Relation Extraction code — Compare
TACRED-Revisited (3 rows) RAG4RE Retrieval-Augmented Generation-based Relation Extraction code — Compare
BioRED (2 rows) PubMedBERT BioRED: A Rich Biomedical Relation Extraction Dataset code — Compare
Dataset: Relationship extraction for knowledge graph creation from biomedical literature (Gene-Disease relationships) (2 rows) DistilBERT Comparison of biomedical relationship extraction methods and... — — Compare
FewRel (2 rows) RoCORE A Relation-Oriented Clustering Method for Open Relation Extraction code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
NYT24 (2 rows) WDec Effective Modeling of Encoder-Decoder Architecture for Joint... code — Compare
REBEL (2 rows) Relik-Large ReLiK: Retrieve and LinK, Fast and Accurate Entity Linking and... code — Compare
SemEval-2010 Task 8 (2 rows) LLM-QA4RE (XXLarge) Aligning Instruction Tasks Unlocks Large Language Models as... code — Compare
WLPC (2 rows) SpanRel Generalizing Natural Language Analysis through Span-relation... code — Compare
2010 i2b2/VA (1 row) Spark NLP Deeper Clinical Document Understanding Using Relation Extraction code — Compare
2012 i2b2 Temporal Relations (1 row) Spark NLP Deeper Clinical Document Understanding Using Relation Extraction code — Compare
2018 n2c2 posology (1 row) Spark NLP Deeper Clinical Document Understanding Using Relation Extraction code — Compare
ADE Corpus (1 row) PFN A Partition Filter Network for Joint Entity and Relation Extraction code Syntology official (archive's flag): 5 ran · 5 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified Compare
Dataset: Relationship extraction for knowledge graph creation from biomedical literature (Gene-Disease relationships) n (1 row) DistilBERT Comparison of biomedical relationship extraction methods and... — — Compare
DuIE (1 row) BiTT A Bidirectional Tree Tagging Scheme for Joint Medical Relation Extraction — — Compare
Google RE (1 row) RELA Sequence Generation with Label Augmentation for Relation Extraction code — Compare
JNLPBA (1 row) SciBERT (SciVocab) SciBERT: A Pretrained Language Model for Scientific Text code — Compare
LPSC-contains (1 row) Stacked_LinkedBERT Exploiting Unary Relations with Stacked Learning for Relation Extraction code — Compare
LPSC-hasproperty (1 row) Stacked_LinkedBERT Exploiting Unary Relations with Stacked Learning for Relation Extraction code — Compare
MUC6 (1 row) iDepNN Neural Relation Extraction Within and Across Sentence Boundaries code — Compare
PGR (1 row) Spark NLP Deeper Clinical Document Understanding Using Relation Extraction code — Compare
sciERC-sent (1 row) RELA Sequence Generation with Label Augmentation for Relation Extraction code — Compare
Wikipedia-Wikidata relations (1 row) ContextAtt Context-Aware Representations for Knowledge Base Relation Extraction code — Compare
WNUT 2020 (1 row) Baseline WNUT-2020 Task 1 Overview: Extracting Entities and Relations from... code — Compare
Babelscape/rebel-dataset (0 rows) no rows in the archive — —

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

82 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 82 until expanded.

Subtasks archive 2025-07-28

15 subtasks in the archive's task tree.

Papers archive 2025-07-28

archive papers tagged: 1,977 · with a code link: 735 · where Syntology ran a sample: 111 (93 with a run with no instrument failure, 18 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (111 of 1,977 tagged: 93 with a run with no instrument failure, 18 where every run was a failure of Syntology's instrument)

Page 1 of 20: papers 1 to 100 of 1,977, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 35 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

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