Deep Reinforcement Learning for Mention-Ranking Coreference Models
Not on the map. Why?
Abstract (arXiv)
Coreference resolution systems are typically trained with heuristic loss functions that require careful tuning. In this paper we instead apply reinforcement learning to directly optimize a neural mention-ranking model for coreference evaluation metrics. We experiment with two approaches: the REINFORCE policy gradient algorithm and a reward-rescaled max-margin objective. We find the latter to be more effective, resulting in significant improvements over the current state-of-the-art on the English and Chinese portions of the CoNLL 2012 Shared Task.
Key facts. Cited by 63 papers in the map (data through 7 Oct 2026). First posted on arXiv 27 Sep 2016.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 29; 2022: 14; 2023: 10; 2024: 6; 2025: 3; 2026: 1.
63 papers.
Sort:
By area:
- 2021GenIE: Generative Information Extractioncited by 53
- 2023VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolutioncited by 52
- 2024CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreferencecited by 51
- 2021ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Historical News Collectionscited by 30
- 2021Automatic Story Generation: Challenges and Attemptscited by 28
- 2022Open-world Story Generation with Structured Knowledge Enhancement: A Comprehensive Surveycited by 21
- 2021Moving on from OntoNotes: Coreference Resolution Model Transfercited by 18
- 2021OntoGUM: Evaluating Contextualized SOTA Coreference Resolution on 12 More Genrescited by 18
- 2021Towards Socially Intelligent Agents with Mental State Transition and Human Utilitycited by 14
- 2021CREAD: Combined Resolution of Ellipses and Anaphora in Dialoguescited by 13
- 2022Machine Reading, Fast and Slow: When Do Models "Understand" Language?cited by 13
- 2021Sequential Cross-Document Coreference Resolutioncited by 12
38 more
- 2021To Protect and To Serve? Analyzing Entity-Centric Framing of Police Violencecited by 12
- 2022Discovering Differences in the Representation of People using Contextualized Semantic Axescited by 12
- 2023AmbiCoref: Evaluating Human and Model Sensitivity to Ambiguous Coreferencecited by 12
- 2021Learning Reasoning Paths over Semantic Graphs for Video-grounded Dialoguescited by 11
- 2021VGNMN: Video-grounded Neural Module Network to Video-Grounded Language Taskscited by 11
- 2021A Survey on Deep Reinforcement Learning for Data Processing and Analyticscited by 10
- 2022DEArt: Dataset of European Artcited by 9
- 2024[Lions: 1] and [Tigers: 2] and [Bears: 3], Oh My! Literary Coreference Annotation with LLMscited by 8
- 2021Reinforcement Learning based Collective Entity Alignment with Adaptive Featurescited by 7
- 2021End-to-end Neural Coreference Resolution Revisited: A Simple yet Effective Baselinecited by 7
- 2022The Dialog Must Go On: Improving Visual Dialog via Generative Self-Trainingcited by 7
- 2023FRAMM: Fair Ranking with Missing Modalities for Clinical Trial Site Selectioncited by 6
- 2021Proto: A Neural Cocktail for Generating Appealing Conversationscited by 5
- 2021Better Call the Plumber: Orchestrating Dynamic Information Extraction Pipelinescited by 4
- 2021Can we Fix the Scope for Coreference? Problems and Solutions for Benchmarks beyond OntoNotescited by 4
- 2022Stylized Knowledge-Grounded Dialogue Generation via Disentangled Template Rewritingcited by 4
- 2022The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources in Natural Language Understanding Systemscited by 4
- 2023A Case Study for Compliance as Code with Graphs and Language Models: Public release of the Regulatory Knowledge Graphcited by 4
- 2021Stay Together: A System for Single and Split-antecedent Anaphora Resolutioncited by 3
- 2021Anatomy of OntoGUM--Adapting GUM to the OntoNotes Scheme to Evaluate Robustness of SOTA Coreference Algorithmscited by 3
- 2025Coreference Resolution for Vietnamese Narrative Textscited by 3
- 2026See Further, Think Deeper: Advancing VLM's Reasoning Ability with Low-level Visual Cues and Reflectioncited by 3
- 2021Situation and Behavior Understanding by Trope Detection on Filmscited by 2
- 2021A Coarse to Fine Question Answering System based on Reinforcement Learningcited by 2
- 2021C3: Compositional Counterfactual Contrastive Learning for Video-grounded Dialoguescited by 2
- 2022Knowledge is Power: Understanding Causality Makes Legal judgment Prediction Models More Generalizable and Robustcited by 2
- 2022VER: Unifying Verbalizing Entities and Relationscited by 2
- 2023PIE-QG: Paraphrased Information Extraction for Unsupervised Question Generation from Small Corporacited by 2
- 2021Secure Bayesian Federated Analytics for Privacy-Preserving Trend Detectioncited by 1
- 2021MAIR: Framework for mining relationships between research articles, strategies, and regulations in the field of explainable artificial intelligencecited by 1
- 2021Knowledge-Grounded Dialogue with Reward-Driven Knowledge Selectioncited by 1
- 2021Newsalyze: Effective Communication of Person-Targeting Biases in News Articlescited by 1
- 2022Hybrid Rule-Neural Coreference Resolution System based on Actor-Critic Learningcited by 1
- 2023Spelling convention sensitivity in neural language modelscited by 1
- 2023Radar de Parit'e: An NLP system to measure gender representation in French news storiescited by 1
- 2023Towards Transparency in Coreference Resolution: A Quantum-Inspired Approachcited by 1
- 2024Asking and Answering Questions to Extract Event-Argument Structurescited by 1
- 2021Coreference Resolution: Are the eliminated spans totally worthless?cited by 0
50 most cited of 63 All 63 (JSON)
- 2026See Further, Think Deeper: Advancing VLM's Reasoning Ability with Low-level Visual Cues and Reflection27 Apr 2026
- 2025BioCoref: Benchmarking Biomedical Coreference Resolution with LLMs29 Oct 2025
- 2025CLAP: Coreference-Linked Augmentation for Passage Retrieval9 Aug 2025
- 2025Coreference Resolution for Vietnamese Narrative Texts28 Apr 2025
- 2024CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreference25 Jun 2024
- 2024Persian Pronoun Resolution: Leveraging Neural Networks and Language Models17 May 2024
- 2024Asking and Answering Questions to Extract Event-Argument Structures25 Apr 2024
- 2024SPLICE: A Singleton-Enhanced PipeLIne for Coreference REsolution25 Mar 2024
- 2024[Lions: 1] and [Tigers: 2] and [Bears: 3], Oh My! Literary Coreference Annotation with LLMs31 Jan 2024
- 2024SPECTRUM: Speaker-Enhanced Pre-Training for Long Dialogue Summarization31 Jan 2024
- 2023Towards Transparency in Coreference Resolution: A Quantum-Inspired Approach1 Dec 2023
- 2023Athena 2.0: Discourse and User Modeling in Open Domain Dialogue3 Aug 2023
38 more
- 2023VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolution21 Jun 2023
- 2023FRAMM: Fair Ranking with Missing Modalities for Clinical Trial Site Selection30 May 2023
- 2023Radar de Parit'e: An NLP system to measure gender representation in French news stories19 Apr 2023
- 2023Bridging Natural Language Processing and Psycholinguistics: computationally grounded semantic similarity datasets for Basque and Spanish19 Apr 2023
- 2023Spelling convention sensitivity in neural language models6 Mar 2023
- 2023A Case Study for Compliance as Code with Graphs and Language Models: Public release of the Regulatory Knowledge Graph3 Feb 2023
- 2023AmbiCoref: Evaluating Human and Model Sensitivity to Ambiguous Coreference1 Feb 2023
- 2023PIE-QG: Paraphrased Information Extraction for Unsupervised Question Generation from Small Corpora3 Jan 2023
- 2022Hybrid Rule-Neural Coreference Resolution System based on Actor-Critic Learning20 Dec 2022
- 2022The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources in Natural Language Understanding Systems15 Dec 2022
- 2022Open-world Story Generation with Structured Knowledge Enhancement: A Comprehensive Survey9 Dec 2022
- 2022VER: Unifying Verbalizing Entities and Relations20 Nov 2022
- 2022Review of coreference resolution in English and Persian8 Nov 2022
- 2022Knowledge is Power: Understanding Causality Makes Legal judgment Prediction Models More Generalizable and Robust6 Nov 2022
- 2022Event and Entity Extraction from Generated Video Captions5 Nov 2022
- 2022DEArt: Dataset of European Art2 Nov 2022
- 2022Discovering Differences in the Representation of People using Contextualized Semantic Axes21 Oct 2022
- 2022Machine Reading, Fast and Slow: When Do Models "Understand" Language?15 Sep 2022
- 2022The Dialog Must Go On: Improving Visual Dialog via Generative Self-Training25 May 2022
- 2022Large-Scale Multi-Document Summarization with Information Extraction and Compression1 May 2022
- 2022Stylized Knowledge-Grounded Dialogue Generation via Disentangled Template Rewriting12 Apr 2022
- 2022Measuring Non-Probabilistic Uncertainty: A cognitive, logical and computational assessment of known and unknown unknowns15 Jan 2022
- 2021Can we Fix the Scope for Coreference? Problems and Solutions for Benchmarks beyond OntoNotes17 Dec 2021
- 2021GenIE: Generative Information Extraction15 Dec 2021
- 2021Newsalyze: Effective Communication of Person-Targeting Biases in News Articles18 Oct 2021
- 2021Tracing Origins: Coreference-aware Machine Reading Comprehension15 Oct 2021
- 2021Anatomy of OntoGUM--Adapting GUM to the OntoNotes Scheme to Evaluate Robustness of SOTA Coreference Algorithms12 Oct 2021
- 2021To Protect and To Serve? Analyzing Entity-Centric Framing of Police Violence11 Sep 2021
- 2021ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Historical News Collections8 Sep 2021
- 2021Proto: A Neural Cocktail for Generating Appealing Conversations6 Sep 2021
- 2021Knowledge-Grounded Dialogue with Reward-Driven Knowledge Selection31 Aug 2021
- 2021A Survey on Deep Reinforcement Learning for Data Processing and Analytics10 Aug 2021
- 2021MAIR: Framework for mining relationships between research articles, strategies, and regulations in the field of explainable artificial intelligence29 Jul 2021
- 2021Secure Bayesian Federated Analytics for Privacy-Preserving Trend Detection28 Jul 2021
- 2021End-to-end Neural Coreference Resolution Revisited: A Simple yet Effective Baseline4 Jul 2021
- 2021C3: Compositional Counterfactual Contrastive Learning for Video-grounded Dialogues16 Jun 2021
- 2021OntoGUM: Evaluating Contextualized SOTA Coreference Resolution on 12 More Genres2 Jun 2021
- 2021A Coarse to Fine Question Answering System based on Reinforcement Learning1 Jun 2021
50 newest of 63 All 63 (JSON)
- 2021GenIE: Generative Information Extractioncited by 53
- 2021ArchivalQA: A Large-scale Benchmark Dataset for Open Domain Question Answering over Historical News Collectionscited by 30
- 2021Automatic Story Generation: Challenges and Attemptscited by 28
- 2022Open-world Story Generation with Structured Knowledge Enhancement: A Comprehensive Surveycited by 21
- 2021Moving on from OntoNotes: Coreference Resolution Model Transfercited by 18
- 2021OntoGUM: Evaluating Contextualized SOTA Coreference Resolution on 12 More Genrescited by 18
- 2021Towards Socially Intelligent Agents with Mental State Transition and Human Utilitycited by 14
- 2021CREAD: Combined Resolution of Ellipses and Anaphora in Dialoguescited by 13
- 2022Machine Reading, Fast and Slow: When Do Models "Understand" Language?cited by 13
- 2021Sequential Cross-Document Coreference Resolutioncited by 12
- 2021To Protect and To Serve? Analyzing Entity-Centric Framing of Police Violencecited by 12
- 2022Discovering Differences in the Representation of People using Contextualized Semantic Axescited by 12
12 most cited of 60 All 60 (JSON)
- 2025BioCoref: Benchmarking Biomedical Coreference Resolution with LLMs29 Oct 2025
- 2025CLAP: Coreference-Linked Augmentation for Passage Retrieval9 Aug 2025
- 2025Coreference Resolution for Vietnamese Narrative Texts28 Apr 2025
- 2024Persian Pronoun Resolution: Leveraging Neural Networks and Language Models17 May 2024
- 2024Asking and Answering Questions to Extract Event-Argument Structures25 Apr 2024
- 2024SPLICE: A Singleton-Enhanced PipeLIne for Coreference REsolution25 Mar 2024
- 2024[Lions: 1] and [Tigers: 2] and [Bears: 3], Oh My! Literary Coreference Annotation with LLMs31 Jan 2024
- 2024SPECTRUM: Speaker-Enhanced Pre-Training for Long Dialogue Summarization31 Jan 2024
- 2023Towards Transparency in Coreference Resolution: A Quantum-Inspired Approach1 Dec 2023
- 2023Athena 2.0: Discourse and User Modeling in Open Domain Dialogue3 Aug 2023
- 2023FRAMM: Fair Ranking with Missing Modalities for Clinical Trial Site Selection30 May 2023
- 2023Radar de Parit'e: An NLP system to measure gender representation in French news stories19 Apr 2023
12 newest of 60 All 60 (JSON)