{"work":{"id":"7ed6cdaa-ed67-4db4-aceb-b7e1b0e6e7c4","openalex_id":null,"doi":null,"arxiv_id":"2109.01652","raw_key":null,"title":"Finetuned Language Models Are Zero-Shot Learners","authors":null,"authors_text":"Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester","year":2021,"venue":"cs.CL","abstract":"This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.\n  We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and surpasses zero-shot 175B GPT-3 on 20 of 25 tasks that we evaluate. FLAN even outperforms few-shot GPT-3 by a large margin on ANLI, RTE, BoolQ, AI2-ARC, OpenbookQA, and StoryCloze. Ablation studies reveal that number of finetuning datasets, model scale, and natural language instructions are key to the success of instruction tuning.","external_url":"https://arxiv.org/abs/2109.01652","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-06-29T09:33:16.738800+00:00","pith_arxiv_id":"2109.01652","created_at":"2026-05-09T05:55:31.196689+00:00","updated_at":"2026-06-29T09:33:16.738800+00:00","title_quality_ok":true,"display_title":"Finetuned Language Models Are Zero-Shot Learners","render_title":"Finetuned Language Models Are Zero-Shot Learners"},"hub":{"state":{"work_id":"7ed6cdaa-ed67-4db4-aceb-b7e1b0e6e7c4","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":138,"external_cited_by_count":null,"distinct_field_count":14,"first_pith_cited_at":"2021-12-01T22:24:34+00:00","last_pith_cited_at":"2026-06-14T20:41:59+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-06-29T11:38:39.091818+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":33},{"context_role":"other","n":5},{"context_role":"dataset","n":3},{"context_role":"method","n":3}],"polarity_counts":[{"context_polarity":"background","n":30},{"context_polarity":"unclear","n":7},{"context_polarity":"use_dataset","n":3},{"context_polarity":"support","n":2},{"context_polarity":"use_method","n":2}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Finetuned Language Models Are Zero-Shot Learners","claims":[{"claim_text":"This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.\n  We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and sur","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"arXiv preprint arXiv:2506.01939, 2025. [53] Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-instruct: Aligning language models with self-generated instruc- tions. InProceedings of the 61st annual meeting of the association for computational linguistics (volume 1: long papers), pages 13484-13508, 2023. [54] Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. Fine","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Self-instruct: Aligning language model with self generated instruc- tions, 2022. [45] Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Anjana Arunkumar, Arjun Ashok, Arut Selvan Dhanasekaran, Atharva Naik, David Stap, et al. Super-naturalinstructions:generalization via declarative instructions on 1600+ tasks. In EMNLP, 2022. [46] Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. Finetuned l","claim_type":"dataset","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Progressively optimized bi-granular document representation for scalable embedding based retrieval. In Proceedings of the ACM Web Conference 2022 . 286-296. [58] Shitao Xiao, Zheng Liu, Yingxia Shao, and Zhao Cao. 2023. RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models. arXiv preprint arXiv:2305.02564 (2023). [59] Shitao Xiao, Zheng Liu, Yingxia Shao, Defu Lian, and Xing Xie. 2021. Matching- oriented product quantization for ad-hoc retrieval.arXiv preprin","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"data synthesis methods. While synthetic data has demonstrated its potential across both small- and large-scale LMs for a variety of general and specialized tasks, including code generation, it also poses several challenges that must be addressed. These challenges include a lack of data diversity [280], the need to ensure the factuality and fidelity of the information [256, 281], and the potential to amplify existing biases or introduce new ones [24, 89]. 5.3 Pre-Training 5.3.1 Model Architecture","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"puter Vision (IJCV) , 2016. https://link.springer.com/article/10.1007/ s11263-014-0748-y . [102] Xiang Zhang, Junbo Zhao, and Yann LeCun. Character-level convolutional net- works for text classification. In Advances in Neural Information Processing Sys- tems (NeurIPS), 2015. https://proceedings.neurips.cc/paper/2015/file/ 250cf8b51c773f3f8dc8b4be867a9a02-Paper.pdf. [103] Ruiqi Zhong, Kristy Lee, Zheng Zhang, and Dan Klein. Adapting language models for zero- shot learning by meta-tuning on dataset","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"luckily guess the correct answer without reliably inducing the underlying rule. We thus identify this inductive gap as a core bottleneck of multimodal ICL. To overcome this fundamental inductive gap, we proposeMMInduction, a framework that instructs multimodal ICL as a structured inductive-deductive process. At its core lies an Inductive Chain- of-Thought (CoT) [42], a reasoning template that guides the model to first analyze the provided cases, extract a generalizable rule, and then apply that ","claim_type":"method","confidence":0.85,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Finetuned Language Models Are Zero-Shot Learners because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (21 contexts).","role_counts":[{"n":21,"context_role":"background"},{"n":2,"context_role":"dataset"},{"n":2,"context_role":"method"},{"n":1,"context_role":"other"}]},"error":null,"updated_at":"2026-05-18T20:51:18.032451+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"cb0687c7-02a4-474c-a3bd-ba5a6e0d00f9","orcid":null,"display_name":"Jason Wei"},{"id":"0778e3ce-2eaf-448e-8c23-5a30ba22c9fb","orcid":null,"display_name":"Maarten Bosma"},{"id":"b5b42d99-0e3b-4be9-a468-9e1b3664f1ba","orcid":null,"display_name":"Vincent Y. Zhao"},{"id":"0878f829-24f6-4ea6-bb35-4c38afcf6802","orcid":null,"display_name":"Kelvin Guu"},{"id":"d0e556f1-bb53-494d-a673-55a5f154e31a","orcid":null,"display_name":"Adams Wei Yu"},{"id":"ab9d7132-eadb-4711-ba7f-4217808bb710","orcid":null,"display_name":"Brian Lester"}]},"error":null,"updated_at":"2026-05-18T20:51:18.019714+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T12:30:34.152538+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":17},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":17},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":16},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":16},{"title":"Training Compute-Optimal Large Language Models","work_id":"b2faf28d-86b7-429c-bc42-469458efc246","shared_citers":15},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":13},{"title":"Measuring Massive Multitask Language Understanding","work_id":"e87ec49a-544b-4ec8-8991-75298c64ff5e","shared_citers":12},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":11},{"title":"LaMDA: Language Models for Dialog Applications","work_id":"1b66d0a5-f6ae-4332-8025-c662dc64b238","shared_citers":11},{"title":"Multitask prompted training enables zero-shot task generalization","work_id":"1857c952-db2f-4810-9e3c-27a862c8d276","shared_citers":11},{"title":"On the Opportunities and Risks of Foundation Models","work_id":"a18039e9-928d-47c9-a836-32656a71bf71","shared_citers":11},{"title":"PaLM: Scaling Language Modeling with Pathways","work_id":"a94f3ef7-2c49-4445-93fe-6ec16aafd966","shared_citers":11},{"title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","work_id":"47ce8be9-e500-407d-af41-ac2d132215eb","shared_citers":11},{"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","shared_citers":10},{"title":"Measuring Mathematical Problem Solving With the MATH Dataset","work_id":"50652ac6-fb7c-4675-a2c2-159c241feb17","shared_citers":10},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":10},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":9},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":9},{"title":"Scaling Instruction-Finetuned Language Models","work_id":"8405abb1-7558-4fdf-af24-f4c52fa77a06","shared_citers":9},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":9},{"title":"Gemini: A Family of Highly Capable Multimodal Models","work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","shared_citers":8},{"title":"Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback","work_id":"a1f2574b-a899-4713-be60-c87ba332656c","shared_citers":8},{"title":"Decoupled Weight Decay Regularization","work_id":"07ef7360-d385-4033-83f7-8384a6325204","shared_citers":7},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":7}],"time_series":[{"n":1,"year":2021},{"n":8,"year":2022},{"n":9,"year":2023},{"n":9,"year":2024},{"n":2,"year":2025},{"n":32,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T12:40:39.876179+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"items":[{"title":"Qwen3 Technical Report","outcome":"unchanged","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"counts":{"fixed":0,"merged":0,"unchanged":1,"quarantined":0,"needs_external_resolution":0},"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-14T12:30:50.005001+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Finetuned Language Models Are Zero-Shot Learners","claims":[{"claim_text":"This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.\n  We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and sur","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"arXiv preprint arXiv:2506.01939, 2025. [53] Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-instruct: Aligning language models with self-generated instruc- tions. InProceedings of the 61st annual meeting of the association for computational linguistics (volume 1: long papers), pages 13484-13508, 2023. [54] Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. Fine","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Self-instruct: Aligning language model with self generated instruc- tions, 2022. [45] Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Anjana Arunkumar, Arjun Ashok, Arut Selvan Dhanasekaran, Atharva Naik, David Stap, et al. Super-naturalinstructions:generalization via declarative instructions on 1600+ tasks. In EMNLP, 2022. [46] Jason Wei, Maarten Bosma, Vincent Y Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M Dai, and Quoc V Le. Finetuned l","claim_type":"dataset","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Progressively optimized bi-granular document representation for scalable embedding based retrieval. In Proceedings of the ACM Web Conference 2022 . 286-296. [58] Shitao Xiao, Zheng Liu, Yingxia Shao, and Zhao Cao. 2023. RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models. arXiv preprint arXiv:2305.02564 (2023). [59] Shitao Xiao, Zheng Liu, Yingxia Shao, Defu Lian, and Xing Xie. 2021. Matching- oriented product quantization for ad-hoc retrieval.arXiv preprin","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"data synthesis methods. While synthetic data has demonstrated its potential across both small- and large-scale LMs for a variety of general and specialized tasks, including code generation, it also poses several challenges that must be addressed. These challenges include a lack of data diversity [280], the need to ensure the factuality and fidelity of the information [256, 281], and the potential to amplify existing biases or introduce new ones [24, 89]. 5.3 Pre-Training 5.3.1 Model Architecture","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"puter Vision (IJCV) , 2016. https://link.springer.com/article/10.1007/ s11263-014-0748-y . [102] Xiang Zhang, Junbo Zhao, and Yann LeCun. Character-level convolutional net- works for text classification. In Advances in Neural Information Processing Sys- tems (NeurIPS), 2015. https://proceedings.neurips.cc/paper/2015/file/ 250cf8b51c773f3f8dc8b4be867a9a02-Paper.pdf. [103] Ruiqi Zhong, Kristy Lee, Zheng Zhang, and Dan Klein. Adapting language models for zero- shot learning by meta-tuning on dataset","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"luckily guess the correct answer without reliably inducing the underlying rule. We thus identify this inductive gap as a core bottleneck of multimodal ICL. To overcome this fundamental inductive gap, we proposeMMInduction, a framework that instructs multimodal ICL as a structured inductive-deductive process. At its core lies an Inductive Chain- of-Thought (CoT) [42], a reasoning template that guides the model to first analyze the provided cases, extract a generalizable rule, and then apply that ","claim_type":"method","confidence":0.85,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Finetuned Language Models Are Zero-Shot Learners because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (21 contexts).","role_counts":[{"n":21,"context_role":"background"},{"n":2,"context_role":"dataset"},{"n":2,"context_role":"method"},{"n":1,"context_role":"other"}]},"error":null,"updated_at":"2026-05-18T20:51:18.027331+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Finetuned Language Models Are Zero-Shot Learners","claims":[{"claim_text":"This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.\n  We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and sur","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Finetuned Language Models Are Zero-Shot Learners because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T12:40:39.787264+00:00"}},"summary":{"title":"Finetuned Language Models Are Zero-Shot Learners","claims":[{"claim_text":"This paper explores a simple method for improving the zero-shot learning abilities of language models. We show that instruction tuning -- finetuning language models on a collection of tasks described via instructions -- substantially improves zero-shot performance on unseen tasks.\n  We take a 137B parameter pretrained language model and instruction-tune it on over 60 NLP tasks verbalized via natural language instruction templates. We evaluate this instruction-tuned model, which we call FLAN, on unseen task types. FLAN substantially improves the performance of its unmodified counterpart and sur","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Finetuned Language Models Are Zero-Shot Learners because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":17},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":17},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":16},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":16},{"title":"Training Compute-Optimal Large Language Models","work_id":"b2faf28d-86b7-429c-bc42-469458efc246","shared_citers":15},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":13},{"title":"Measuring Massive Multitask Language Understanding","work_id":"e87ec49a-544b-4ec8-8991-75298c64ff5e","shared_citers":12},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":11},{"title":"LaMDA: Language Models for Dialog Applications","work_id":"1b66d0a5-f6ae-4332-8025-c662dc64b238","shared_citers":11},{"title":"Multitask prompted training enables zero-shot task generalization","work_id":"1857c952-db2f-4810-9e3c-27a862c8d276","shared_citers":11},{"title":"On the Opportunities and Risks of Foundation Models","work_id":"a18039e9-928d-47c9-a836-32656a71bf71","shared_citers":11},{"title":"PaLM: Scaling Language Modeling with Pathways","work_id":"a94f3ef7-2c49-4445-93fe-6ec16aafd966","shared_citers":11},{"title":"Scaling Language Models: Methods, Analysis & Insights from Training Gopher","work_id":"47ce8be9-e500-407d-af41-ac2d132215eb","shared_citers":11},{"title":"Language Models are Few-Shot Learners","work_id":"214732c0-2edd-44a0-af9e-28184a2b8279","shared_citers":10},{"title":"Measuring Mathematical Problem Solving With the MATH Dataset","work_id":"50652ac6-fb7c-4675-a2c2-159c241feb17","shared_citers":10},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":10},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":9},{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","shared_citers":9},{"title":"Scaling Instruction-Finetuned Language Models","work_id":"8405abb1-7558-4fdf-af24-f4c52fa77a06","shared_citers":9},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":9},{"title":"Gemini: A Family of Highly Capable Multimodal Models","work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","shared_citers":8},{"title":"Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback","work_id":"a1f2574b-a899-4713-be60-c87ba332656c","shared_citers":8},{"title":"Decoupled Weight Decay Regularization","work_id":"07ef7360-d385-4033-83f7-8384a6325204","shared_citers":7},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":7}],"time_series":[{"n":1,"year":2021},{"n":8,"year":2022},{"n":9,"year":2023},{"n":9,"year":2024},{"n":2,"year":2025},{"n":32,"year":2026}],"dependency_candidates":[]},"authors":[{"id":"d0e556f1-bb53-494d-a673-55a5f154e31a","orcid":null,"display_name":"Adams Wei Yu","source":"manual","import_confidence":0.72},{"id":"ab9d7132-eadb-4711-ba7f-4217808bb710","orcid":null,"display_name":"Brian Lester","source":"manual","import_confidence":0.72},{"id":"cb0687c7-02a4-474c-a3bd-ba5a6e0d00f9","orcid":null,"display_name":"Jason Wei","source":"manual","import_confidence":0.72},{"id":"0878f829-24f6-4ea6-bb35-4c38afcf6802","orcid":null,"display_name":"Kelvin Guu","source":"manual","import_confidence":0.72},{"id":"0778e3ce-2eaf-448e-8c23-5a30ba22c9fb","orcid":null,"display_name":"Maarten Bosma","source":"manual","import_confidence":0.72},{"id":"b5b42d99-0e3b-4be9-a468-9e1b3664f1ba","orcid":null,"display_name":"Vincent Y. Zhao","source":"manual","import_confidence":0.72}]}}