Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

MAQInstruct: Instruction-based Unified Event Relation Extraction

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read MAQInstruct recasts event-relation extraction as multiple-answer selection among marked event mentions, using a bipartite matching loss to remove answer-order effects and cutting inference queries from $n^2$ to $k \times n$.

desk verdict Solid efficiency win for instruction-based ERE, but the order-invariance claim only covers the answer list, not the dependency chain that actually drives performance. read the letter →

arxiv 2502.03954 v1 pith:K6SX36NR submitted 2025-02-06 cs.CL cs.AI

Jun Xu , Mengshu Sun , Zhiqiang Zhang , Jun Zhou This is my paper · ORCID
classification cs.CLcs.AI
keywords eventrelationextractioninstructiontuninglargelanguagemodelsmultiple-answerquestionansweringbipartitematchinglossdependencyparsingchaincoreferencetemporal
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper's central claim is that instruction-based event relation extraction fails for two fixable reasons: it asks one question per event pair (so $n^2$ samples per document), and autoregressive generation makes the answer order matter even though event relations are not sequential. MAQInstruct fixes both by inverting the instruction: given a relation, the model selects which marked event mentions in the context satisfy it, cutting the sample count to $k \times n$ where $k$ is the number of relation types. A bipartite matching loss treats the selected answers as an unordered set, so the model is not penalized for producing them in any order. If right, the same framework handles coreference, temporal, causal, and sub-event relations with one instruction-tuned LLM, and on MAVEN-ERE it beats both the pairwise instruction baseline and a classification baseline while reducing coreference inference queries from 631,486 to 17,780 and wall-clock time from about 813 to 25 minutes.

What carries the argument

The central machinery is the multiple-answer question format plus a bipartite matching loss. Each event mention in the context is preceded by a distinct marker token (for example <0x85>), and an instruction names one relation and one query mention; the model must output a dependency parsing chain—a sequence of typed syntactic edges connecting event mentions—followed by the set of marked mentions that satisfy the relation. The bipartite matching loss uses the Hungarian algorithm to align the generated answer list with the gold list under the best permutation, so the order of the answers has no effect on the gradient; the dependency parsing chain is trained with ordinary cross-entropy and provides the structured scaffold that makes scattered event relations recoverable.

What would settle it

Shuffle the order of the edges in the dependency parsing chain at inference time (for example, reverse the chain, or sort edges randomly) while keeping the same answers; if MATRES or MECI F1 drops materially below the reported numbers, the method still depends on generation order, contradicting the claim that the bipartite matching loss removes sequence sensitivity. Alternatively, replace the generated chain with an unordered set of dependency edges; if accuracy collapses, the chain is functioning as a fixed-order crutch rather than a content scaffold.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is that the bottleneck in instruction-based event relation extraction is the question format, not the language model. Instead of enumerating all event pairs and asking 'what relation holds?', MAQInstruct asks 'which marked events stand in relation R to the query event?', with the context carrying the candidate mentions as special marker tokens. The label is a dependency parsing chain followed by a comma-separated list of answers; the loss over the answer list is computed with a Hungarian matching, so any permutation of the generated answers receives the same loss. Experiments show this formulation improves F1 over InstructERE by roughly four points per subtask and over the classification-based BertERE by 0.4 to 1.7 points on MAVEN-ERE, and it transfers zero-shot to HiEve, MATRES, and MECI better than the prior instruction method.

Load-bearing premise

The premise that carries the method is that the model can generate the dependency parsing chain in its fixed order, and that this chain is what makes the scattered event relations recoverable; the ablation shows that without the chain the method loses the most accuracy, so if chain order is itself unreliable the claimed sequence independence is only shifted, not eliminated.

Editorial extensions

If this is right

  • A single instruction-tuned LLM can extract coreference, temporal, causal, and sub-event relations using the same framework, with only the instruction changing.
  • The number of inference samples scales with the number of relation types times the number of mentions rather than the square of the mention count, making document-level extraction with dozens or hundreds of mentions feasible for LLMs.
  • The bipartite matching loss makes answer generation order irrelevant, which the paper shows holds across random, reverse, distance-sorted, and dictionary-sorted answer orders.
  • The dependency parsing chain is the largest accuracy contributor: removing it drops MATRES F1 from 84.7 to 81.4 and MECI F1 from 62.3 to 58.4, so structured syntactic scaffolding is central to the gain.
  • Zero-shot performance on unseen event-relation datasets (HiEve, MATRES, MECI) improves relative to the prior instruction-based method, and general NLU performance is not degraded.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The relation-as-instruction, mentions-as-answer inversion is a general template: any task with few relation or role types and many candidate spans (for example argument role labeling, entity linking, or coreference beyond events) could inherit the same sample-count reduction.
  • Because the dependency parsing chain is itself a generated sequence in a fixed order, the paper's sequence-independence claim is only partially demonstrated; a stress test would shuffle the chain's edges at inference, and if accuracy drops, the ordering problem has moved to the chain rather than disappearing.
  • The efficiency advantage depends on $k \ll n$; on tasks with a large relation inventory, the gap between $k \times n$ and $n^2$ closes, so the method is most compelling in sparse-relation regimes.
  • The ablation shows markers alone carry most of the gain when the chain is present, suggesting that marker-anchored decoding, rather than the chain's linguistic content per se, might be the active ingredient; a test would be to replace the dependency parse with a random but consistent structured scaffold.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes MAQInstruct, an instruction-tuned framework for unified event relation extraction. It replaces pairwise event-event questions with relation-specific instructions that ask the model to select one or more event mentions from the context, reducing the number of inference queries from quadratic to linear in the number of mentions per document. It also introduces a bipartite matching loss (BPM) applied to the multiple-answer portion of the generated output, intended to make the model insensitive to the order in which answers are generated. Experiments on MAVEN-ERE compare MAQInstruct with BertERE and an InstructERE baseline across three LLMs (ChatGLM3, Qwen, Llama2), and cross-corpus/zero-shot results are reported on HiEve, MATRES, and MECI. The paper reports consistent F1 improvements over InstructERE (about 3.9-4.2 points on the four MAVEN-ERE subtasks for Llama2) and an inference-time reduction from 813 to 25 minutes for coreference on an A100-80G setup.

Significance. The empirical scope is a genuine strength: the method is evaluated on four ERE datasets, three LLM backbones, and five seeds, and the inference-cost comparison in Table 2 is concrete and central. The main efficiency claim (quadratic to linear query count) is well motivated and supported by the reported costs. If the order-invariance claim were fully established, the BPM component would be a useful and transferable idea. However, the present evidence supports order-invariance only for the final answer list, not for the full generated output, which weakens the contribution's stated scope.

major comments (3)
  1. [§2.2, Eq. (4); §3.6, Table 3] The claim that MAQInstruct reduces the dependency on the generation sequence is only partially supported. The BPM loss in Eq. (4) is applied only to the multiple-answer part; the dependency parsing chain (DPC) is generated with ordinary cross-entropy under a fixed order (Appendix A), and Table 3 shows that removing the DPC causes the largest performance drop (MATRES F1 84.7 to 81.4; MECI F1 62.3 to 58.4). No experiment varies the DPC order, although Appendix E itself attributes errors to 'the complexity of the dependency parsing chain.' The paper should either restrict the order-invariance claim to the multiple-answer list or demonstrate that the DPC is order-stable (e.g., by comparing parser-determined, shuffled, and reversed DPC orders).
  2. [Table 1; Appendix B] The comparison against BertERE is presented as a positive result, but the gains are as small as 0.4 and at most 1.7 F1 points, and no standard deviations or significance tests are reported even though Appendix B states the results are averaged over five random seeds. Without error bars or a paired test, the claim that MAQInstruct 'enhances' over BertERE is not statistically supported; at minimum, report per-seed variance or a significance test for the four MAVEN-ERE subtasks.
  3. [§2.2, Eqs. (2)-(3)] The formal definition of the matching loss is ambiguous. N2 is introduced as the length of the multiple-answer part in tokens, but the permutation Ψ_N2 in Eqs. (2)-(3) is over token positions, with c_i a target vocabulary id; for multi-token answers, matching individual tokens is not equivalent to matching whole answer spans. Please state clearly whether the Hungarian algorithm operates over answer spans or tokens, and how variable-length answers are handled; this is necessary to reproduce the loss exactly.
minor comments (5)
  1. [Section 3.1] InstructERE is never cited or referenced; please either cite the original work or state explicitly that it is an internal baseline, and if it is based on the authors' prior ChatUIE [37], say so.
  2. [Section 3.2] The claimed '4.9%' improvement for coreference over InstructERE does not match any of the four coreference metrics in Table 1 (the closest is 4.2 points on BLANC); please correct the number.
  3. [Figure 3] The plot has no axis labels or legend entries explaining what is being measured; please add labels for F1 and for the dataset/sequence-condition grouping.
  4. [Section 3.5] The term 'zero-shot' is used for models trained on MAVEN-ERE and evaluated on HiEve/MATRES/MECI; this is cross-corpus transfer rather than zero-shot learning in the usual sense, so the terminology should be clarified.
  5. [Table 2; Section 1] The n×n sample count should specify that n is the number of event mentions per document rather than the total number of mentions in the dataset; otherwise the reported 631,486 coreference queries for InstructERE appear inconsistent with the 17,780 total mentions.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the reported gains are measured on external benchmarks and the bipartite matching loss is an externally established set-generation technique applied to ERE, not a fitted input renamed as a prediction.

full rationale

MAQInstruct's central claims are empirical: F1 improvements on MAVEN-ERE, HiEve, MATRES, and MECI, an inference-time reduction from explicit query counts, and ablation comparisons in Table 3. None of these numbers is obtained by fitting a parameter to a quantity and then reporting that same quantity as a prediction. The k×n versus n×n sample reduction is an arithmetic consequence of the task reformulation rather than a discovered result, and the paper does not disguise it as a prediction. The bipartite matching loss is imported from the externally cited set-generation literature (OTSeq2Set [3], One2Set [40]), where permutation-invariance is a known property of the Hungarian matching objective; Section 3.4 tests the loss under different answer orderings rather than assuming its effect. The only self-citations ([37], [38]) appear in standard metric and related-work reference lists and are not load-bearing for the paper's mechanism. The dependency-parsing-chain order dependence raised in the skeptical reading is a genuine limitation: the DPC is an ordered generated sequence, and its removal causes the largest ablation drop, so the 'generation sequence' claim strictly covers the multiple-answer block rather than the entire output. But that is a scope/generality gap, not a circular reduction: the DPC is produced by the external CoreNLP parser with deterministic tie-breaking and is evaluated by ablation, so the derivation chain does not assume what it purports to prove. The InstructERE baseline is described without a citation, which is a source-attribution concern, but it is not an example of a result reducing to its own input.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central method introduces no new physical entities nor fitted constants beyond a hand-set loss weight lambda. The main assumptions are about the reliability of the dependency parsing chain, the expressiveness of the QA formulation, and the standard optimal matching algorithm. No parameter is fitted to data to produce the headline results; the reported gains come from the training setup and loss choice.

free parameters (1)
  • lambda (bipartite matching loss weight) = 0.2
    Set by hand (Appendix B); controls the contribution of the BPM loss to the total loss in Eq. 4, and the paper's main claims depend on the BPM loss being effective.
assumptions (4)
  • standard math The Hungarian algorithm finds the optimal permutation for token-level answer alignment.
    Invoked in Eq. 2 to compute the best matching between predicted tokens and target answers; correctness is standard.
  • domain assumption The dependency parser (CoreNLP) produces dependency chains that are valid and useful for ERE.
    The DPC is included in the training label and is shown in ablations to be the most important component; the paper does not evaluate parser errors or their propagation.
  • domain assumption Instruction-tuned LLMs can map relation-specific instructions to correct event selections.
    The method's effectiveness depends on the LLM's ability to follow the constructed instructions, supported only by the reported empirical results.
  • domain assumption The multiple-answer QA formulation can recover all relations from k*n queries without needing pair-wise information.
    The paper assumes that for each relation type and each event mention, asking for all related events covers all valid pairs, but does not analyze recall loss for relations that require context from both mentions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MAQInstruct: Instruction-based Unified Event Relation Extraction." pith.science (2026). https://pith.science/paper/K6SX36NR

@misc{pith2026250203954,
  author       = {Pith},
  title        = {Pith review of: MAQInstruct: Instruction-based Unified Event Relation Extraction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K6SX36NR}},
  note         = {Machine review of arXiv:2502.03954}
}
read the original abstract

Extracting event relations that deviate from known schemas has proven challenging for previous methods based on multi-class classification, MASK prediction, or prototype matching. Recent advancements in large language models have shown impressive performance through instruction tuning. Nevertheless, in the task of event relation extraction, instruction-based methods face several challenges: there are a vast number of inference samples, and the relations between events are non-sequential. To tackle these challenges, we present an improved instruction-based event relation extraction framework named MAQInstruct. Firstly, we transform the task from extracting event relations using given event-event instructions to selecting events using given event-relation instructions, which reduces the number of samples required for inference. Then, by incorporating a bipartite matching loss, we reduce the dependency of the instruction-based method on the generation sequence. Our experimental results demonstrate that MAQInstruct significantly improves the performance of event relation extraction across multiple LLMs.

Figures

Figures reproduced from arXiv: 2502.03954 by the authors.

Figure 1
Figure 1. Different event relation extraction methods. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Cross-entropy loss vs bipartite matching loss. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. The performance of different answer sequences. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: The F1 score of different instruction-based LLMs [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: The performance of different LLMs on natural lan [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]
Figure 6
Figure 6. Figure 6: A, B, D represent event mentions, while C denotes [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: Two examples demonstrating the use of MAQIn [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. GDLLM: A Global Distance-aware Modeling Approach Based on Large Language Models for Event Temporal Relation Extraction

    cs.CL 2025-08 conditional novelty 5.0 of 10

    GDLLM improves event temporal relation extraction by feeding LLM-generated probability distributions into a graph attention network, achieving state-of-the-art micro-F1 scores on TB-Dense and MATRES.

Reference graph

Works this paper leans on

46 extracted references · 43 canonical work pages · cited by 1 Pith paper

  1. [1]

    Amit Bagga and Breck Baldwin. 1998. Algorithms for scoring coreference chains. In The first international conference on language resources and evaluation workshop on linguistics coreference, Vol. 1. 563–566

    work page 1998

  2. [2]

    Shany Barhom, Vered Shwartz, Alon Eirew, Michael Bugert, Nils Reimers, and Ido Dagan. 2019. Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution. In ACL. 4179–4189

    work page 2019

  3. [3]

    Jie Cao and Yin Zhang. 2022. OTSeq2Set: An Optimal Transport Enhanced Sequence-to-Set Model for Extreme Multi-label Text Classification. In EMNLP. 5588–5597

    work page 2022

  4. [4]

    Tommaso Caselli and Piek Vossen. 2017. The Event StoryLine Corpus: A New Benchmark for Causal and Temporal Relation Extraction. In ACL. 77–86

    work page 2017

  5. [5]

    Meiqi Chen, Yubo Ma, Kaitao Song, Yixin Cao, Yan Zhang, and Dongsheng Li

    work page

  6. [6]

    Prafulla Kumar Choubey and Ruihong Huang. 2017. Event Coreference Reso- lution by Iteratively Unfolding Inter-dependencies among Events. In EMNLP. 2124–2133

    work page 2017

  7. [7]

    Shiyao Cui, Jiawei Sheng, Xin Cong, Quangang Li, Tingwen Liu, and Jinqiao Shi

    work page

  8. [8]

    Goran Glavas, Jan Snajder, Marie-Francine Moens, and Parisa Kordjamshidi. 2014. HiEve: A Corpus for Extracting Event Hierarchies from News Stories. In LREC. 3678–3683

    work page 2014

Show all 46 references
  1. [9]

    Minghao Hu, Yuxing Peng, Zhen Huang, and Dongsheng Li. 2019. A Multi- Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoning. In EMNLP-IJCNLP. Hong Kong, China

    2019

  2. [10]

    Zhilei Hu, Zixuan Li, Xiaolong Jin, Long Bai, Saiping Guan, Jiafeng Guo, and Xueqi Cheng. 2023. Semantic Structure Enhanced Event Causality Identification. In ACL. 10901–10913

    2023

  3. [11]

    Zhilei Hu, Zixuan Li, Daozhu Xu, Long Bai, Cheng Jin, Xiaolong Jin, Jiafeng Guo, and Xueqi Cheng. 2023. ProtoEM: A Prototype-Enhanced Matching Framework for Event Relation Extraction. arXiv:2309.12892 [cs.CL]

    2023 arXiv

  4. [12]

    Quzhe Huang, Yutong Hu, Shengqi Zhu, Yansong Feng, Chang Liu, and Dongyan Zhao. 2023. More than Classification: A Unified Framework for Event Temporal Relation Extraction. In ACL. 9631–9646

    2023

  5. [13]

    EunJeong Hwang, Jay-Yoon Lee, Tianyi Yang, Dhruvesh Patel, Dongxu Zhang, and Andrew McCallum. 2022. Event-Event Relation Extraction using Probabilistic Box Embedding. In ACL. 235–244

    2022

  6. [14]

    Viet Dac Lai, Amir Pouran Ben Veyseh, Minh Van Nguyen, Franck Dernoncourt, and Thien Huu Nguyen. 2022. MECI: A Multilingual Dataset for Event Causality Identification. In COLING. 2346–2356

    2022

  7. [15]

    Jing Lu and Vincent Ng. 2021. Constrained Multi-Task Learning for Event Coreference Resolution. In NAACL-HLT. 4504–4514. MAQInstruct: Instruction-based Unified Event Relation Extraction WWW Companion ’25, April 28-May 2, 2025, Sydney, NSW, Australia

    2021

  8. [16]

    Yaojie Lu, Qing Liu, Dai Dai, Xinyan Xiao, Hongyu Lin, Xianpei Han, Le Sun, and Hua Wu. 2022. Unified Structure Generation for Universal Information Extraction. In ACL. Dublin, Ireland, 5755–5772

    2022

  9. [17]

    Xiaoqiang Luo. 2005. On coreference resolution performance metrics. In HLT (Vancouver, British Columbia, Canada). Morristown, NJ, USA, 25–32

    2005

  10. [18]

    Hieu Man, Nghia Trung Ngo, Linh Ngo Van, and Thien Huu Nguyen. 2022. Selecting Optimal Context Sentences for Event-Event Relation Extraction. In AAAI. 11058–11066

    2022

  11. [19]

    Minh Van Nguyen, Bonan Min, Franck Dernoncourt, and Thien Nguyen. 2022. Learning Cross-Task Dependencies for Joint Extraction of Entities, Events, Event Arguments, and Relations. In EMNLP. 9349–9360

    2022

  12. [20]

    Minh Van Nguyen, Bonan Min, Franck Dernoncourt, and Thien Huu Nguyen

  13. [21]

    Qiang Ning, Hao Wu, and Dan Roth. 2018. A Multi-Axis Annotation Scheme for Event Temporal Relations. In ACL. 1318–1328

    2018

  14. [22]

    RECASENS and E

    M. RECASENS and E. HOVY. 2011. BLANC: Implementing the Rand index for coreference evaluation. Natural Language Engineering 17, 4 (2011), 485–510

    2011

  15. [23]

    In NAACL

    Joint Extraction of Entities, Relations, and Events via Modeling Inter- Instance and Inter-Label Dependencies. In NAACL. 4363–4374

  16. [24]

    Shirong Shen, Heng Zhou, Tongtong Wu, and Guilin Qi. 2022. Event Causality Identification via Derivative Prompt Joint Learning. In COLING. 2288–2299

    2022

  17. [25]

    Xingwei Tan, Gabriele Pergola, and Yulan He. 2023. Event Temporal Relation Extraction with Bayesian Translational Model. In EACL. 1117–1130

    2023

  18. [26]

    Elad Segal, Avia Efrat, Mor Shoham, Amir Globerson, and Jonathan Berant. 2020. A Simple and Effective Model for Answering Multi-span Questions. In EMNLP. 3074–3080

    2020

  19. [27]

    Vilain, John D

    Marc B. Vilain, John D. Burger, John S. Aberdeen, Dennis Connolly, and Lynette Hirschman. 1995. A Model-Theoretic Coreference Scoring Scheme. In Message Understanding Conference

    1995

  20. [28]

    Somin Wadhwa, Silvio Amir, and Byron C. Wallace. 2023. Revisiting Relation Extraction in the era of Large Language Models. In ACL. 15566–15589

    2023

  21. [29]

    Hieu Minh Tran, Duy Phung, and Thien Huu Nguyen. 2021. Exploiting Docu- ment Structures and Cluster Consistencies for Event Coreference Resolution. In ACL/IJCNLP. 4840–4850

    2021

  22. [30]

    Gardner, Dan Roth, and Muhao Chen

    Haoyu Wang, Hongming Zhang, Yuqian Deng, Jacob R. Gardner, Dan Roth, and Muhao Chen. 2023. Extracting or Guessing? Improving Faithfulness of Event Temporal Relation Extraction. In EACL. 541–553

    2023

  23. [31]

    Xiaozhi Wang, Yulin Chen, Ning Ding, Hao Peng, Zimu Wang, Yankai Lin, Xu Han, Lei Hou, Juanzi Li, Zhiyuan Liu, Peng Li, and Jie Zhou. 2022. MAVEN-ERE: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction. In EMNLP. 926–941

    2022

  24. [32]

    Haoyu Wang, Muhao Chen, Hongming Zhang, and Dan Roth. 2020. Joint Con- strained Learning for Event-Event Relation Extraction. In EMNLP. 696–706

    2020

  25. [33]

    Xiao Wang, Weikang Zhou, Can Zu, Han Xia, Tianze Chen, Yuansen Zhang, Rui Zheng, Junjie Ye, Qi Zhang, Tao Gui, Jihua Kang, Jingsheng Yang, Siyuan Li, and Chunsai Du. 2023. InstructUIE: Multi-task Instruction Tuning for Unified Information Extraction. arXiv:2304.08085 [cs.CL]

    2023 arXiv

  26. [34]

    Haoyang Wen and Heng Ji. 2021. Utilizing Relative Event Time to Enhance Event-Event Temporal Relation Extraction. In EMNLP. 10431–10437

    2021

  27. [35]

    Xiaozhi Wang, Hao Peng, Yong Guan, Kaisheng Zeng, Jianhui Chen, Lei Hou, Xu Han, Yankai Lin, Zhiyuan Liu, Ruobing Xie, Jie Zhou, and Juanzi Li. 2024. MAVEN-ARG: Completing the Puzzle of All-in-One Event Understanding Dataset with Event Argument Annotation. In ACL. 4072–4091

    2024

  28. [36]

    Xinglin Xiao, Yijie Wang, Nan Xu, Yuqi Wang, Hanxuan Yang, Minzheng Wang, Yin Luo, Lei Wang, Wenji Mao, and Daniel Zeng. 2024. YAYI-UIE: A Chat- Enhanced Instruction Tuning Framework for Universal Information Extraction. arXiv:2312.15548 [cs.CL]

    2024 arXiv

  29. [37]

    Jun Xu, Mengshu Sun, Zhiqiang Zhang, and Jun Zhou. 2024. ChatUIE: Exploring Chat-based Unified Information Extraction Using Large Language Models. In LREC/COLING 2024. 3146–3152

    2024

  30. [38]

    Wei Xiang, Chuanhong Zhan, and Bang Wang. 2023. DAPrompt: Determin- istic Assumption Prompt Learning for Event Causality Identification. CoRR abs/2307.09813 (2023)

    2023 arXiv

  31. [39]

    Deming Ye, Yankai Lin, Peng Li, and Maosong Sun. 2022. Packed Levitated Marker for Entity and Relation Extraction. In ACL. 4904–4917

    2022

  32. [40]

    Jiacheng Ye, Tao Gui, Yichao Luo, Yige Xu, and Qi Zhang. 2021. One2Set: Gener- ating Diverse Keyphrases as a Set. In ACL/IJCNLP. 4598–4608

    2021

  33. [41]

    Jun Xu, Weidi Xu, Mengshu Sun, Taifeng Wang, and Wei Chu. 2022. Extracting Trigger-sharing Events via an Event Matrix. In EMNLP. Abu Dhabi, United Arab Emirates, 1189–1201

    2022

  34. [42]

    Yutao Zeng, Xiaolong Jin, Saiping Guan, Jiafeng Guo, and Xueqi Cheng. 2020. Event Coreference Resolution with their Paraphrases and Argument-aware Em- beddings. In COLING. 3084–3094

    2020

  35. [43]

    " with

    Jie Zhou, Shenpo Dong, Hongkui Tu, Xiaodong Wang, and Yong Dou. 2022. RSGT: Relational Structure Guided Temporal Relation Extraction. In COLING. 2001–2010. A Dependency Parsing Chain We use the Stanford NLP toolkit’s CoreNLP Dependency Parser to create a dependency parse tree ...

    2022

  36. [44]

    Changsen Yuan, Heyan Huang, Yixin Cao, and Yonggang Wen. 2023. Discrimi- native Reasoning with Sparse Event Representation for Document-level Event- Event Relation Extraction. In ACL. 16222–16234

    2023

  37. [2022]

    In COLING

    Event Causality Extraction with Event Argument Correlations. In COLING. 2300–2312

  38. [2024]

    Improving Large Language Models in Event Relation Logical Prediction. In ACL. 9451–9478

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.