{"work":{"id":"1ff75a14-302e-4906-aa86-1f96fbcf12ff","openalex_id":"https://openalex.org/W4417459816","doi":"10.48550/arxiv.2512.13564","arxiv_id":"2512.13564","raw_key":null,"title":"Memory in the Age of AI Agents","authors":null,"authors_text":"Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang, Boyang Liu, Fangyi Zhu","year":2025,"venue":"cs.CL","abstract":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of contemporary agent memory systems. This work aims to provide an up-to-date landscape of current agent memory research. We begin by clearly delineating the scope of agent memory and distinguishing it from related concepts such as LLM memory, retrieval augmented generation (RAG), and context engineering. We then examine agent memory through the unified lenses of forms, functions, and dynamics. From the perspective of forms, we identify three dominant realizations of agent memory, namely token-level, parametric, and latent memory. From the perspective of functions, we propose a finer-grained taxonomy that distinguishes factual, experiential, and working memory. From the perspective of dynamics, we analyze how memory is formed, evolved, and retrieved over time. To support practical development, we compile a comprehensive summary of memory benchmarks and open-source frameworks. Beyond consolidation, we articulate a forward-looking perspective on emerging research frontiers, including memory automation, reinforcement learning integration, multimodal memory, multi-agent memory, and trustworthiness issues. We hope this survey serves not only as a reference for existing work, but also as a conceptual foundation for rethinking memory as a first-class primitive in the design of future agentic intelligence.","external_url":"https://arxiv.org/abs/2512.13564","cited_by_count":1,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2512.13564","created_at":"2026-05-09T05:55:29.459949+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":false,"display_title":"Memory in the Age of AI Agents","render_title":"Memory in the Age of AI Agents"},"hub":{"state":{"work_id":"1ff75a14-302e-4906-aa86-1f96fbcf12ff","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":104,"external_cited_by_count":1,"distinct_field_count":13,"first_pith_cited_at":"2026-01-13T14:48:34+00:00","last_pith_cited_at":"2026-07-09T01:15:03+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":"2026-08-17T21:09:28.523623+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":15},{"context_role":"baseline","n":2}],"polarity_counts":[{"context_polarity":"background","n":14},{"context_polarity":"baseline","n":2},{"context_polarity":"unclear","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Memory in the Age of AI Agents","claims":[{"claim_text":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of co","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"ral and spatial grounding, long-video efficiency, memory- augmented systems, agentic video understanding, stream- ing processing, and grounded video reasoning. Existing surveys have provided valuable summaries of video-language understanding, temporal grounding, post- training, memory, token reduction, and multimodal rea- soning [28], [29], [30], [31], [32], [33], [34], [35]. For ex- ample, Nguyen et al. [28] review video-language under- standing from model architecture, training, and data per- ","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"1 Information-theoretic intuition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 A.2 Overview of existing memory benchmarks: LoCoMo and LongMemEval . . . . . 31 2 1 The memory problem in AI systems 1.1 Memory query types exceed what semantic retrieval can support The term \"memory\" covers several mechanisms in AI systems, and this paper addresses only one of them [ 1, 2].Parametric memoryis knowledge encoded in model weights through training or fine-tuning: it is implicit, hard to ins","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Best results are highlighted in bold, and second-best results are underlined . Model Method Single-hop Multi-hop Open-domain Temporal OverallGPT-4o-mini [4] Full Context 45.89 25.01 15.49 23.83 35.57 RAG [13] 52.19 32.17 23.21 30.77 42.25 Mem0 [7] 47.65 38.72 28.64 48.93 45.10 MemoryOS [16] 48.62 35.27 20.02 41.15 42.84 HippoRAG [10] 54.84 33.59 28.59 48.17 47.92 A-Mem [37] 44.65 27.02 12.14 45.85 39.65 CAM [15] 50.58 33.55 18.23 44.14 44.10 CompassMem [11] 57.36 38.84 26.61 57.96 52.18 MemCoT 6","claim_type":"baseline","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Moreover, existing evidence suggests that ostensibly deleted knowledge may still be recoverable through intermediate-layer traces or paraphrasing attacks [34]. In contrast, replacing private RAG contexts with fully synthetic data [48], or reducing sensitive corpus exposure through privacy-preserving vector databases and data processing frameworks [14], can indeed help mitigate the risk that models regurgitate pre-existing private data. Nevertheless, they still struggle to adequately cover a more","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"},{"claim_text":"on LLM agents via a lightweight DSL, validated on embodied and autonomous driving tasks. RoboSafe [485] prevents implicit risks in VLM-driven agents via hybrid reasoning with executable safety logic. 6.2 Memory Agent memory (episodic logs, RAG corpora, and conversation histories) enables cross-session experience accumulation but creates a durable attack surface for both integrity and confidentiality violations [136,245, 392, 461]. The OWASP Top 10 for Agentic Applications classifies memory poiso","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"},{"claim_text":"cently, ActorBreaker [32] investigates multi-turn jailbreaks through natural distribution shifts, constructing attack trajectories that ex- ploit semantic relationships between benign and toxic concepts. By modeling interactions among different actors in a dialogue, Actor- Breaker demonstrates that aligned LLMs may exhibit safety gaps under naturally evolving multi-turn conversations. 2.2 LLM Memory LLM memory [14, 22, 48] research investigates how to preserve and reuse information beyond a sing","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Memory in the Age of AI Agents because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (15 contexts).","role_counts":[{"n":15,"context_role":"background"},{"n":2,"context_role":"baseline"}]},"error":null,"updated_at":"2026-07-04T06:46:37.978514+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"c1bfa7d9-045b-41d2-9e7e-17607f6829f2","orcid":null,"display_name":"Yuyang Hu"},{"id":"3a57a690-faa8-4cea-9cfa-8ae0590eafbe","orcid":null,"display_name":"Shichun Liu"},{"id":"f6f0b866-55ec-4906-97b7-5065ae129ce9","orcid":null,"display_name":"Yanwei Yue"},{"id":"3266248f-b1a6-4906-8ed8-5e019625e66e","orcid":null,"display_name":"Guibin Zhang"},{"id":"65cb7958-421e-4727-b713-122a9544bc7e","orcid":null,"display_name":"Boyang Liu"},{"id":"5f4f7c16-9bbe-4f7b-820d-82ec61ca1aaa","orcid":null,"display_name":"Fangyi Zhu"}]},"error":null,"updated_at":"2026-07-04T06:46:37.976019+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T18:09:54.287669+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory","work_id":"a5aed26c-a248-48b6-a59e-f7693fcb180a","shared_citers":20},{"title":"A-MEM: Agentic Memory for LLM Agents","work_id":"3b98feb2-fdb1-479a-bbe4-2c298a4592e2","shared_citers":16},{"title":"MemGPT: Towards LLMs as Operating Systems","work_id":"2698f5ad-c84c-40ca-b839-0912dae10ba2","shared_citers":16},{"title":"Zep: A Temporal Knowledge Graph Architecture for Agent Memory","work_id":"515c933e-12ae-439d-a7ff-c07fee482dfb","shared_citers":11},{"title":"LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory","work_id":"9074870f-aee7-4103-b167-ac6473a8a9b3","shared_citers":8},{"title":"Memos: A memory os for ai system","work_id":"e686f058-6c5c-4484-a88e-26d945657ed3","shared_citers":8},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":8},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":7},{"title":"From Local to Global: A Graph RAG Approach to Query-Focused Summarization","work_id":"588618d7-fd41-4053-b34d-a981f8793039","shared_citers":7},{"title":"Evaluating Very Long-Term Conversational Memory of LLM Agents","work_id":"2d8c9fb7-9ace-4925-8fe1-f4e44625d04c","shared_citers":6},{"title":"Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning","work_id":"2a070440-2167-4398-be97-c2d4c3ee3541","shared_citers":6},{"title":"ReAct: Synergizing Reasoning and Acting in Language Models","work_id":"407a2351-25f1-497d-b611-f77d0292a8e6","shared_citers":6},{"title":"Lightmem: Lightweight and efficient memory-augmented generation","work_id":"154bcc76-21c8-4f79-87d4-0b9cfdda70c0","shared_citers":5},{"title":"What Deserves Memory: Adaptive Memory Distillation for LLM Agents","work_id":"8025a606-5218-4c5a-b99f-83049e7feac5","shared_citers":5},{"title":"Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents","work_id":"4519e90e-26c9-43ef-a89b-c8e8a8378280","shared_citers":4},{"title":"arXiv:2405.14831 [cs.CL] https://arxiv.org/abs/2405.14831","work_id":"c964b1e8-cbf5-49dd-ac34-b8f733b02a9d","shared_citers":4},{"title":"Bernstein","work_id":"b5bf85fe-4fb7-4966-b0b2-9ccf9d3b11b9","shared_citers":4},{"title":"Evaluating memory in llm agents via incremental multi-turn interactions","work_id":"df5096c3-e738-472b-953c-fb510ee9e68d","shared_citers":4},{"title":"Evaluating very long-term conversational memory of LLM agents","work_id":"6bbf9aa9-b06f-489e-8977-aeb1261f2ef8","shared_citers":4},{"title":"MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents","work_id":"9217d063-8ac8-4b58-bc7e-b977719660c2","shared_citers":4},{"title":"Memos: An operating system for memory-augmented generation (mag) in large language models","work_id":"62f802d2-2c22-4140-9b3d-b8b7220f772e","shared_citers":4},{"title":"MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents","work_id":"64baa7f3-849b-485e-b081-2074d82f1364","shared_citers":4},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":4},{"title":"Retrieval-Augmented Generation for Large Language Models: A Survey","work_id":"b80d2790-6cd9-4c87-b3c4-de404f99a80e","shared_citers":4}],"time_series":[{"n":34,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T18:09:54.318216+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":"2026-05-14T18:09:36.056388+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Memory in the Age of AI Agents","claims":[{"claim_text":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of co","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"ral and spatial grounding, long-video efficiency, memory- augmented systems, agentic video understanding, stream- ing processing, and grounded video reasoning. Existing surveys have provided valuable summaries of video-language understanding, temporal grounding, post- training, memory, token reduction, and multimodal rea- soning [28], [29], [30], [31], [32], [33], [34], [35]. For ex- ample, Nguyen et al. [28] review video-language under- standing from model architecture, training, and data per- ","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"1 Information-theoretic intuition . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 A.2 Overview of existing memory benchmarks: LoCoMo and LongMemEval . . . . . 31 2 1 The memory problem in AI systems 1.1 Memory query types exceed what semantic retrieval can support The term \"memory\" covers several mechanisms in AI systems, and this paper addresses only one of them [ 1, 2].Parametric memoryis knowledge encoded in model weights through training or fine-tuning: it is implicit, hard to ins","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Best results are highlighted in bold, and second-best results are underlined . Model Method Single-hop Multi-hop Open-domain Temporal OverallGPT-4o-mini [4] Full Context 45.89 25.01 15.49 23.83 35.57 RAG [13] 52.19 32.17 23.21 30.77 42.25 Mem0 [7] 47.65 38.72 28.64 48.93 45.10 MemoryOS [16] 48.62 35.27 20.02 41.15 42.84 HippoRAG [10] 54.84 33.59 28.59 48.17 47.92 A-Mem [37] 44.65 27.02 12.14 45.85 39.65 CAM [15] 50.58 33.55 18.23 44.14 44.10 CompassMem [11] 57.36 38.84 26.61 57.96 52.18 MemCoT 6","claim_type":"baseline","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Moreover, existing evidence suggests that ostensibly deleted knowledge may still be recoverable through intermediate-layer traces or paraphrasing attacks [34]. In contrast, replacing private RAG contexts with fully synthetic data [48], or reducing sensitive corpus exposure through privacy-preserving vector databases and data processing frameworks [14], can indeed help mitigate the risk that models regurgitate pre-existing private data. Nevertheless, they still struggle to adequately cover a more","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"},{"claim_text":"on LLM agents via a lightweight DSL, validated on embodied and autonomous driving tasks. RoboSafe [485] prevents implicit risks in VLM-driven agents via hybrid reasoning with executable safety logic. 6.2 Memory Agent memory (episodic logs, RAG corpora, and conversation histories) enables cross-session experience accumulation but creates a durable attack surface for both integrity and confidentiality violations [136,245, 392, 461]. The OWASP Top 10 for Agentic Applications classifies memory poiso","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"},{"claim_text":"cently, ActorBreaker [32] investigates multi-turn jailbreaks through natural distribution shifts, constructing attack trajectories that ex- ploit semantic relationships between benign and toxic concepts. By modeling interactions among different actors in a dialogue, Actor- Breaker demonstrates that aligned LLMs may exhibit safety gaps under naturally evolving multi-turn conversations. 2.2 LLM Memory LLM memory [14, 22, 48] research investigates how to preserve and reuse information beyond a sing","claim_type":"background","confidence":0.85,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks Memory in the Age of AI Agents because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (15 contexts).","role_counts":[{"n":15,"context_role":"background"},{"n":2,"context_role":"baseline"}]},"error":null,"updated_at":"2026-07-04T06:46:37.592416+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Memory in the Age of AI Agents","claims":[{"claim_text":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of co","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Memory in the Age of AI Agents because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T18:09:39.657507+00:00"}},"summary":{"title":"Memory in the Age of AI Agents","claims":[{"claim_text":"Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of co","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Memory in the Age of AI Agents because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory","work_id":"a5aed26c-a248-48b6-a59e-f7693fcb180a","shared_citers":20},{"title":"A-MEM: Agentic Memory for LLM Agents","work_id":"3b98feb2-fdb1-479a-bbe4-2c298a4592e2","shared_citers":16},{"title":"MemGPT: Towards LLMs as Operating Systems","work_id":"2698f5ad-c84c-40ca-b839-0912dae10ba2","shared_citers":16},{"title":"Zep: A Temporal Knowledge Graph Architecture for Agent Memory","work_id":"515c933e-12ae-439d-a7ff-c07fee482dfb","shared_citers":11},{"title":"LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory","work_id":"9074870f-aee7-4103-b167-ac6473a8a9b3","shared_citers":8},{"title":"Memos: A memory os for ai system","work_id":"e686f058-6c5c-4484-a88e-26d945657ed3","shared_citers":8},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":8},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":7},{"title":"From Local to Global: A Graph RAG Approach to Query-Focused Summarization","work_id":"588618d7-fd41-4053-b34d-a981f8793039","shared_citers":7},{"title":"Evaluating Very Long-Term Conversational Memory of LLM Agents","work_id":"2d8c9fb7-9ace-4925-8fe1-f4e44625d04c","shared_citers":6},{"title":"Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning","work_id":"2a070440-2167-4398-be97-c2d4c3ee3541","shared_citers":6},{"title":"ReAct: Synergizing Reasoning and Acting in Language Models","work_id":"407a2351-25f1-497d-b611-f77d0292a8e6","shared_citers":6},{"title":"Lightmem: Lightweight and efficient memory-augmented generation","work_id":"154bcc76-21c8-4f79-87d4-0b9cfdda70c0","shared_citers":5},{"title":"What Deserves Memory: Adaptive Memory Distillation for LLM Agents","work_id":"8025a606-5218-4c5a-b99f-83049e7feac5","shared_citers":5},{"title":"Agentic Memory: Learning Unified Long-Term and Short-Term Memory Management for Large Language Model Agents","work_id":"4519e90e-26c9-43ef-a89b-c8e8a8378280","shared_citers":4},{"title":"arXiv:2405.14831 [cs.CL] https://arxiv.org/abs/2405.14831","work_id":"c964b1e8-cbf5-49dd-ac34-b8f733b02a9d","shared_citers":4},{"title":"Bernstein","work_id":"b5bf85fe-4fb7-4966-b0b2-9ccf9d3b11b9","shared_citers":4},{"title":"Evaluating memory in llm agents via incremental multi-turn interactions","work_id":"df5096c3-e738-472b-953c-fb510ee9e68d","shared_citers":4},{"title":"Evaluating very long-term conversational memory of LLM agents","work_id":"6bbf9aa9-b06f-489e-8977-aeb1261f2ef8","shared_citers":4},{"title":"MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents","work_id":"9217d063-8ac8-4b58-bc7e-b977719660c2","shared_citers":4},{"title":"Memos: An operating system for memory-augmented generation (mag) in large language models","work_id":"62f802d2-2c22-4140-9b3d-b8b7220f772e","shared_citers":4},{"title":"MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents","work_id":"64baa7f3-849b-485e-b081-2074d82f1364","shared_citers":4},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":4},{"title":"Retrieval-Augmented Generation for Large Language Models: A Survey","work_id":"b80d2790-6cd9-4c87-b3c4-de404f99a80e","shared_citers":4}],"time_series":[{"n":34,"year":2026}],"dependency_candidates":[]},"authors":[{"id":"65cb7958-421e-4727-b713-122a9544bc7e","orcid":null,"display_name":"Boyang Liu","source":"manual","import_confidence":0.72},{"id":"5f4f7c16-9bbe-4f7b-820d-82ec61ca1aaa","orcid":null,"display_name":"Fangyi Zhu","source":"manual","import_confidence":0.72},{"id":"3266248f-b1a6-4906-8ed8-5e019625e66e","orcid":null,"display_name":"Guibin Zhang","source":"manual","import_confidence":0.72},{"id":"3a57a690-faa8-4cea-9cfa-8ae0590eafbe","orcid":null,"display_name":"Shichun Liu","source":"manual","import_confidence":0.72},{"id":"f6f0b866-55ec-4906-97b7-5065ae129ce9","orcid":null,"display_name":"Yanwei Yue","source":"manual","import_confidence":0.72},{"id":"c1bfa7d9-045b-41d2-9e7e-17607f6829f2","orcid":null,"display_name":"Yuyang Hu","source":"manual","import_confidence":0.72}]}}