ODDA: An OODA-Driven Diverse Data Augmentation Framework for Low-Resource Relation Extraction
Open accessAbstract
Data Augmentation (DA) has emerged as a promising solution to address the scarcity of high-quality annotated data in low-resource relation extraction (LRE).Leveraging large language models (LLMs), DA has significantly improved the performance of RE models with considerably fewer parameters.However, existing DA methods struggle with diversity misalignments, as they neglect the diversity required by the model and generate homogeneous augmentations that do not cover the inter-sample and inter-relation variability, leading to suboptimal performance.Inspired by the Observe-Orient-Decide-Act (OODA) framework, which provides a robust theoretical foundation for iterative decision-making under dynamic conditions, we propose an OODA-driven Diverse DA method (ODDA), guiding the data generation and selection process.ODDA first observes the RE model's behavior to select effective demonstrations for LLMs.Next, it orients LLMs towards generating diverse data by replacing schema constraints with attribute constraints.Then ODDA decides on the final augmented dataset with overall diversity from a global search and finally acts to train the RE model.Extensive experiments on three widely-used benchmarks demonstrate that ODDA consistently outperforms state-of-the-art baselines, achieving average F1 improvements of 3.1% across various LRE scenarios while maintaining enhanced model stability.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Existing long-term memory methods typically follow two directions: continuously recording, summarizing, and updating dialogue histories as memory, or transforming conversations into structured graph-based memory using OpenIE.
“These approaches typically follow two directions: (i) continuously recording, summarizing, and updating dialogue histories as memory [2, 23, 39, 44], or (ii) transforming conversations into structured graphbased memory using techniques such as OpenIE [19, 40, 57, 58].”
From HingeMem: Boundary Guided Long-Term Memory with Query Adaptive Retrieval for Scalable Dialogues (2026) · cites this paper as [57]
The Related Work section discusses data augmentation for low-resource relation extraction, not methods for maintaining long-term dialogue memory. Searches for recording and summarizing dialogue histories and for OpenIE-based graph memory found no passage addressing either direction in our copy. (LLM, conf=0.77)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
BLEU: a method for automatic evaluation of machine translation
Papineni, K., Roukos, S., Ward, T., & Zhu, W.-J. (2001) · Proceedings of the 40th Annual Meeting on Association for Computational Linguistics - ACL '02 · doi:10.3115/1073083.1073135 · Cited by 21,344 · see citing works
Secretan, P. (2020) · Encyclopedia of the UN Sustainable Development Goals · doi:10.1007/978-3-319-95870-5_300146 · Cited by 6,546
EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Tasks
Wei, J., & Zou, K. (2019) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d19-1670 · Cited by 2,242 · cited 1× · see citing works
“…data by substituting tokens, like synonym replacement (Mueller and Thyagarajan, 2016) or token-level operations like random insertion, swap, and deletion (Wei and Zou, 2019), which is further enhanced by leveraging word embeddings to generate contextually similar replacement (Jiao et al., 2020).”
A Diversity-Promoting Objective Function for Neural Conversation Models
Li, J., Galley, M., Brockett, C., Gao, J., & Dolan, B. (2016) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/n16-1014 · Cited by 1,981 · cited 8× · see citing works
“Following the previous work (Zheng et al., 2024), we introduce the Distinct (Li et al., 2016) to assess the diversity of synthetic sentences.”
InfluentialTinyBERT: Distilling BERT for natural language understanding
Jiao, X., Yin, Y., Shang, L., Jiang, X., Chen, X., Li, L., Wang, F., & Liu, Q. (2020) · Findings of the Association for Computational Linguistics: EMNLP 2020 · doi:10.18653/v1/2020.findings-emnlp.372 · Cited by 1,786 · cited 7× · see citing works
“…data by substituting tokens, like synonym replacement (Mueller and Thyagarajan, 2016) or token-level operations like random insertion, swap, and deletion (Wei and Zou, 2019), which is further enhanced by leveraging word embeddings to generate contextually similar replacement (Jiao et al., 2020).”
InfluentialPosition-aware Attention and Supervised Data Improve Slot Filling
Zhang, Y., Zhong, V., Chen, D., Angeli, G., & Manning, C. D. (2017) · Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1004 · Cited by 1,009 · cited 5× · see citing works
“Figure 11 illustrates the RE performances at various scaling factors for ReTACRED and SemEval datasets.”
InfluentialSiamese Recurrent Architectures for Learning Sentence Similarity
Mueller, J., & Thyagarajan, A. (2016) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v30i1.10350 · Cited by 935 · cited 7× · see citing works
“(1) WordNet Synonym Substitution (WSS) (Mueller and Thyagarajan, 2016): re-placing tokens with synonyms from WordNet; (2) Word Embedding Substitution (WES) (Jiao et al., 2020): replacing tokens with contextual embed-dings from BERT; (3) LAMBADA (Anaby-Tavor et al., 2020): fine-tuning generative models to generate cand…”
InfluentialSemEval-2010 Task 8: Multi-Way Classification of Semantic Relations Between Pairs of Nominals
Hendrickx, I., Kim, S., Kozareva, Z., Nakov, P., Séaghdha, D. Ó., Padó, S., Pennacchiotti, M., Romano, L., & Szpakowicz, S. (2009) · SEW@NAACL-HLT · doi:10.3115/1621969.1621986 · Cited by 929 · see citing works
How Variable May a Constant be? Measures of Lexical Richness in Perspective
Tweedie, F. J., & Baayen, R. H. (1998) · Computers and the Humanities · doi:10.1023/a:1001749303137 · Cited by 432 · cited 9× · see citing works
“TTR measures the ratio of the number of different words to the total number of words in the dependency path between the given head entity and tail entity.”
InfluentialTexygen: A Benchmarking Platform for Text Generation Models
Zhu, Y., Lu, S., Zheng, L., Guo, J., Zhang, W., Wang, J., & Yu, Y. (2018) · The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval · doi:10.1145/3209978.3210080 · Cited by 408 · cited 2× · see citing works
“For lexical diversity, we employ Type-Token Ratio (TTR) (Tweedie and Baayen, 1998), Distinct-N (Li et al., 2016), and Self-BLEU (Zhu et al., 2018), while semantic diversity is measured using Average Pairwise sample Similarity (APS) (Mishra et al., 2020) by computing inter-class and intra-class APS…”
Large Language Model as Attributed Training Data Generator: A Tale of Diversity and BiasPreprint
Yu, Y., Zhuang, Y.-C., Zhang, J., Meng, Y., Ratner, A. J., Krishna, R., Shen, J., & Zhang, C. (2023) · Neural Information Processing Systems · doi:10.48550/arxiv.2306.15895 · Cited by 396 · cited 9× · see citing works
“The effectiveness of DA heavily depends on its quality and diversity, which plays a pivotal role in RE performance (Yu et al., 2023), particularly in capturing complex relation patterns and semantic variations.”
InfluentialKnowPrompt: Knowledge-aware Prompt-tuning with Synergistic Optimization for Relation Extraction
Chen, X., Zhang, N., Xie, X., Deng, S., Yao, Y., Tan, C., Huang, F., Si, L., & Chen, H. (2022) · Proceedings of the ACM Web Conference 2022 · doi:10.1145/3485447.3511998 · Cited by 389 · cited 2× · see citing works
“KnowPrompt (Chen et al., 2022) is a knowledge-aware prompt-tuning framework for RE.”
SemEval-2010 Task 8: Multi-Way Classification of Semantic Relations between Pairs of NominalsPreprint
Hendrickx, I., Kim, S., Kozareva, Z., Nakov, P., Séaghdha, D. Ó., Padó, S., Pennacchiotti, M., Romano, L., & Szpakowicz, S. (2010) · International Workshop on Semantic Evaluation · doi:10.48550/arxiv.1911.10422 · Cited by 367 · see citing works
Do Not Have Enough Data? Deep Learning to the Rescue!
Anaby-Tavor, A., Carmeli, B., Goldbraich, E., Kantor, A., Kour, G., Shlomov, S., Tepper, N., & Zwerdling, N. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6233 · Cited by 352 · cited 8× · see citing works
“(1) WordNet Synonym Substitution (WSS) (Mueller and Thyagarajan, 2016): re-placing tokens with synonyms from WordNet; (2) Word Embedding Substitution (WES) (Jiao et al., 2020): replacing tokens with contextual embed-dings from BERT; (3) LAMBADA (Anaby-Tavor et al., 2020): fine-tuning generative models to generate cand…”
InfluentialLi, B., Fang, G., Yang, Y., Wang, Q.-S., Ye, W., Zhao, W., & Zhang, S.-K. (2023) · arXiv.org · doi:10.48550/arxiv.2304.11633 · Cited by 223 · cited 1× · see citing works
“Despite Large Language Models (LLMs) possessing zero/few-shot capabilities, they still struggle on RE tasks (Li et al., 2023a; Han et al., 2023) with rich pre-defined patterns, complex classification spaces (Xu et al., 2023b), and high computational costs.”
An Improved Baseline for Sentence-level Relation Extraction
Zhou, W., & Chen, M. (2021) · AACL · doi:10.18653/v1/2022.aacl-short.21 · Cited by 162 · cited 2× · see citing works
“(1) TYPMarker (Zhou and Chen, 2022): a fine-tuning method that leverages entity typed markers; (2) KnowPrompt (Chen et al., 2022): a prompt-tuning method that uses knowledge-aware continuous tuning with synergistic optimization.”
Syntactic Data Augmentation Increases Robustness to Inference Heuristics
Min, J., McCoy, R. T., Das, D., Pitler, E., & Linzen, T. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.212 · Cited by 153 · cited 1× · see citing works
“Traditional DA methods (Cai et al., 2020; Min et al., 2020), relying on surface-level operations like token replacement ad deletion, often yield suboptimal data quality and limited diversity.”
ZeroGen: Efficient Zero-shot Learning via Dataset Generation
Ye, J., Gao, J., Li, Q., Xu, H., Feng, J., Wu, Z., Yu, C., & Kong, L. (2022) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2022.emnlp-main.801 · Cited by 143 · cited 1× · see citing works
“Previous research (Ye et al., 2022; Xu et al., 2023b) indicates that generating more data may not improve RE per-LAMBADA formance.”
WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation
Liu, A., Swayamdipta, S., Smith, N. A., & Choi, Y. (2022) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2022.findings-emnlp.508 · Cited by 131 · cited 1× · see citing works
“To ensure these attributes are both rational and comprehensive, we employ a human-ai collaboration scheme (Liu et al., 2022a; Wiegreffe et al., 2022) to select the most suitable, high-quality attributes for the RE task.”
MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence FrontiersPreprint
Pillutla, K., Swayamdipta, S., Zellers, R., Thickstun, J., Welleck, S., Choi, Y., & Harchaoui, Z. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2102.01454 · Cited by 128 · cited 8× · see citing works
“Beyond sufficient diversity, we also evaluate sample quality using MAUVE ⋆ and Front-Integral ⋆ (Pillutla et al., 2023).”
InfluentialReframing Human-AI Collaboration for Generating Free-Text Explanations
Wiegreffe, S., Hessel, J., Swayamdipta, S., Riedl, M. O., & Choi, Y. (2022) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2022.naacl-main.47 · Cited by 97 · cited 1× · see citing works
“To ensure these attributes are both rational and comprehensive, we employ a human-ai collaboration scheme (Liu et al., 2022a; Wiegreffe et al., 2022) to select the most suitable, high-quality attributes for the RE task.”
How to Unleash the Power of Large Language Models for Few-shot Relation Extraction?
Xu, X., Zhu, Y.-Q., Wang, X.-H., & Zhang, N. (2023) · SUSTAINLP · doi:10.48550/arxiv.2305.01555 · Cited by 86 · cited 9×
“Subsequent work enhances data diversity through structured guidance, incorporating schema (Xu et al., 2023b) and keywords (Zheng et al., 2024) into LLM-based generation.”
InfluentialFabbri, A. R., Han, S., Li, H., Li, H., Ghazvininejad, M., Joty, S., Radev, D., & Mehdad, Y. (2021) · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT) · doi:10.18653/v1/2021.naacl-main.57 · Cited by 75 · cited 1× · see citing works
“While these methods lack diversity, later work like Back-translation (Fabbri et al., 2021), masked language modeling (Lowell et al., 2021), and GPT-2 fine-tuning (Anaby-Tavor et al., 2020) utilize generative models but struggle with relation consistency (Chen, 2024).”
Semi-supervised Relation Extraction via Incremental Meta Self-Training
Hu, X., Zhang, C., Ma, F., Liu, C., Wen, L., & Yu, P. S. (2021) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2021.findings-emnlp.44 · Cited by 69 · cited 1× · see citing works
“Various approaches are explored in recent years to address LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al.,…”
Cai, H., Chen, H., Song, Y., Zhang, C., Zhao, X., & Yin, D. (2020) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2020.acl-main.564 · Cited by 61 · cited 1× · see citing works
“Traditional DA methods (Cai et al., 2020; Min et al., 2020), relying on surface-level operations like token replacement ad deletion, often yield suboptimal data quality and limited diversity.”
An Empirical Study on Information Extraction using Large Language Models
Han, R., Yang, C., Peng, T., Tiwari, P., Wan, X., Liu, L., & Wang, B. (2023) · arXiv · doi:10.48550/arxiv.2305.14450 · Cited by 52 · see citing works
Synergizing RAG and Reasoning: A Systematic ReviewPreprint
Gao, Y.-F., Xiong, Y., Zhong, Y., et al. (2025) · arXiv.org · doi:10.48550/arxiv.2504.15909 · Cited by 48 · cited 1× · see citing works
“Various approaches are explored in recent years to address LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al.,…”
Self-citationS2ynRE: Two-stage Self-training with Synthetic data for Low-resource Relation Extraction
Xu, B., Wang, Q., Lyu, Y., et al. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2023.acl-long.455 · Cited by 44 · cited 1×
“As DA methods are data-centric, we keep the RE baseline models simple yet competitive (Xu et al., 2023a).”
Pre-training to Match for Unified Low-shot Relation Extraction
Liu, F., Lin, H.-Y., Han, X.-P., Cao, B., & Sun, L. (2022) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2203.12274 · Cited by 41 · cited 1×
“…approaches are explored in recent years to address LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al.,…”
Re-TACRED: Addressing Shortcomings of the TACRED Dataset
Stoica, G., Platanios, E. A., & Póczos, B. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i15.17631 · Cited by 37 · cited 1× · see citing works
“We conduct experiments on three public relation extraction datasets: TACRED (Zhang et al., 2017), Re-TACRED (Stoica et al., 2021), and SemEval (Hendrickx et al., 2019).”
Unsupervised Data Augmentation with Naive Augmentation and without Unlabeled Data
Lowell, D., Howard, B. E., Lipton, Z. C., & Wallace, B. (2021) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2021.emnlp-main.408 · Cited by 23 · cited 1× · see citing works
“While these methods lack diversity, later work like Back-translation (Fabbri et al., 2021), masked language modeling (Lowell et al., 2021), and GPT-2 fine-tuning (Anaby-Tavor et al., 2020) utilize generative models but struggle with relation consistency (Chen, 2024).”
GDA: Generative Data Augmentation Techniques for Relation Extraction Tasks
Hu, X., Liu, A., Tan, Z., et al. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2305.16663 · Cited by 23 · cited 4×
“GDA (Hu et al., 2023) advances this field by introducing generative models and establishing diversity as a crucial consideration in DA.”
InfluentialSelf-citationMa, M. D., Wang, X., Kung, P.-N., Brantingham, P. J., Peng, N., & Wang, W. W. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i17.29839 · Cited by 22 · cited 1× · see citing works
“By leveraging LLMs’ ability to perform the inverse task of generating sentences from triples (Ma et al., 2024), DA forms a bridge between LLMs and RE models, ultimately boosting LRE performance.”
Liu, L., Pillutla, K., Welleck, S., et al. (2021) · Neural Information Processing Systems · doi:10.48550/arxiv.2106.07898 · Cited by 22 · cited 5×
“Beyond sufficient diversity, we also evaluate sample quality using MAUVE ⋆ and Front-Integral ⋆ (Pillutla et al., 2023).”
InfluentialInformation Extraction in Low-Resource Scenarios: Survey and Perspective
Deng, S., Ma, Y., Zhang, N., Cao, Y., & Hooi, B. (2024) · IEEE International Conference on Knowledge Graph (ICKG) · doi:10.1109/ickg63256.2024.00013 · Cited by 21 · cited 8× · see citing works
“We carefully analyze the diversity and quality of the augmented data and experiments validate that ODDA produces more diverse data and achieves superior performance in LRE scenarios.”
InfluentialLinguistic representations for fewer-shot relation extraction across domains
Gururaja, S., Dutt, R., Liao, T.-G., & Rosé, C. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.48550/arxiv.2307.03823 · Cited by 18 · cited 1×
“Data Augmentation (DA) offers a direct and efficient approach for LRE compared to meta-learning (Veyseh et al., 2023) or transfer learning (Gururaja et al., 2023).”
Towards Realistic Low-resource Relation Extraction: A Benchmark with Empirical Baseline Study
Xu, X., Chen, X., Zhang, N., Xie, X., Chen, X., & Chen, H.-J. (2022) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2210.10678 · Cited by 18 · cited 1×
“In this paper, we follow the previous research (Xu et al., 2022) for k -shot data sampling.”
SynthesizRR: Generating Diverse Datasets with Retrieval Augmentation
Divekar, A., & Durrett, G. (2024) · Conference on Empirical Methods in Natural Language Processing · doi:10.48550/arxiv.2405.10040 · Cited by 17 · cited 1×
“It reveals that the data generated by other methods exhibits clear boundaries with the target distribution and such domain shift may decrease the model performance (Divekar and Durrett, 2024).”
DQI: Measuring Data Quality in NLPPreprint
Mishra, S. R., Arunkumar, A., Sachdeva, B., Bryan, C., & Baral, C. R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.00816 · Cited by 16 · cited 7× · see citing works
“Furthermore, to gain deeper insights, we compute both intra-class APS and inter-class APS, respectively capturing diversity within the same relation type and across different relation types.”
InfluentialCan We Survive without Labelled Data in NLP? Transfer Learning for Open Information Extraction
Sarhan, I., & Spruit, M. R. (2020) · Applied Sciences · doi:10.3390/app10175758 · Cited by 13 · cited 1× · see citing works
“…LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al., 2023b), and data augmentation (Zhang et al., 2024).”
Making LLMs as Fine-Grained Relation Extraction Data Augmentor
Zheng, Y., Ke, W., Liu, Q., et al. (2024) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2024/736 · Cited by 13 · cited 8×
“Following the previous work (Zheng et al., 2024), we introduce the Distinct (Li et al., 2016) to assess the diversity of synthetic sentences.”
InfluentialGenerating Labeled Data for Relation Extraction: A Meta Learning Approach with Joint GPT-2 Training
Veyseh, A. P. B., Dernoncourt, F., Min, B., et al. (2023) · Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2023.findings-acl.727 · Cited by 5 · cited 2×
“…in recent years to address LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al., 2023b), and data…”
Effective Tool Augmented Multi-Agent Framework for Data Analysis
Zhang, X., Mao, Z., Chen, Z., & Gao, S. (2024) · Data Intelligence · doi:10.3724/2096-7004.di.2024.0013 · Cited by 4 · cited 1× · see citing works
“…LRE challenges (Gao et al., 2025; Oida-Onesa and Ballera, 2024): meta-learning (Hu et al., 2021; Liu et al., 2022b; Veyseh et al., 2023), transfer learning (Sarhan and Spruit, 2020; Gu-ruraja et al., 2023), instruction prompting (Li et al., 2023b), and data augmentation (Zhang et al., 2024).”
Self-citationMeta-PKE: Memory-Enhanced Task-Adaptive Personal Knowledge Extraction in Daily Life
Zhong, Y., Wu, F., Guo, M., Zhang, X., Wang, M., & Wang, H. (2025) · Information Processing & Management · doi:10.1016/j.ipm.2025.104097 · Cited by 3 · see citing works
Self-citation
Flagged first, then by citations
7 of the 51 references couldn’t be matched to a known paper.
Research works printed but not linked: 9
- Huajun Chen. 2024. Large knowledge model: Perspec- tives and challenges. Data Intelligence, 6(3):587- 620.
- Alexander R. Fabbri, Simeng Han, Haoyuan Li, Hao- ran Li, Marjan Ghazvininejad, Shafiq R. Joty, Dragomir R. Radev, and Yashar Mehdad. 2021. Im- proving zero and few-shot abstractive summariza- tion with intermediate fine-tuning and data augmenta- tion. In NAACL-HLT, pages 704-717. Association for Computational Linguistics.
- Guozheng Li, Peng Wang, and Wenjun Ke. 2023b. Re- visiting large language models as zero-shot relation extractors. In EMNLP (Findings), pages 6877-6892. Association for Computational Linguistics.
- Francesco Molfese, Simone Conia, Riccardo Orlando, and Roberto Navigli. 2024. ZEBRA: zero-shot example-based retrieval augmentation for common- sense question answering. In EMNLP, pages 22429- 22444. Association for Computational Linguistics.
- Rosel Oida-Onesa and Melvin A. Ballera. 2024. Fine tuning language models: A tale of two low-resource languages. Data Intelligence, 6(4):946-967.
- ACL.
- Krishna Pillutla, Lang Liu, John Thickstun, Sean Welleck, Swabha Swayamdipta, Rowan Zellers, Se- woong Oh, Yejin Choi, and Zaid Harchaoui. 2023. MAUVE Scores for Generative Models: Theory and Practice. JMLR.
- Swabha Swayamdipta, Roy Schwartz, Nicholas Lourie, Yizhong Wang, Hannaneh Hajishirzi, Noah A. Smith, and Yejin Choi. 2020. Dataset cartography: Map- ping and diagnosing datasets with training dynamics. In EMNLP (1), pages 9275-9293. Association for Computational Linguistics.
- "He is being charged with murder," the detective stated during the press conference.
Cited by
Sorted by citations
HingeMem: Boundary Guided Long-Term Memory with Query Adaptive Retrieval for Scalable Dialogues
Zhong, Y., Gao, Y., & Wang, H. (2026) · Proceedings of the ACM Web Conference 2026 · doi:10.1145/3774904.3792089 · Cited by 1 · see citing works
Self-citationRetrieval-Augmented Relation Extraction for Medical Knowledge Graphs
Song, Y., Li, X., Li, Y., Li, Y., Wu, K., Li, Y., & Zhang, K. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-7299-1_20
Wang, S., Che, Y., Jia, X., et al. (2026) · Electronics · doi:10.3390/electronics15122684
- Authors
- Zhong, Y., Gao, Y., Zhang, X., & Wang, H.
- Affiliations
- Tongji University (CN)
- Huawei Technologies (China) (CN)
- Venue
- Findings of the Association for Computational Linguistics: ACL 2025
- Published
- 2025
- Pages
- 267-285
- OpenAlex
- W4412888960
- Cited by
- 3
- References
- 51
- Citations by year
- 2026: 3
- FWCI
- 2.84
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Findings of the Association for Computational Linguistics: ACL (Findings of ACL) (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Graph Neural Networks
- Keywords
- data augmentation, large language models, schema constraints, attribute constraints, demonstration selection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
