{"work":{"id":"56b6b58d-e73a-4317-896e-36ac5f84e957","openalex_id":null,"doi":null,"arxiv_id":"2408.06292","raw_key":null,"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","authors":null,"authors_text":"Chris Lu, Cong Lu, Robert Tjarko Lange, Jakob Foerster, Jeff Clune, David Ha","year":2024,"venue":"cs.AI","abstract":"One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which generates novel research ideas, writes code, executes experiments, visualizes results, describes its findings by writing a full scientific paper, and then runs a simulated review process for evaluation. In principle, this process can be repeated to iteratively develop ideas in an open-ended fashion, acting like the human scientific community. We demonstrate its versatility by applying it to three distinct subfields of machine learning: diffusion modeling, transformer-based language modeling, and learning dynamics. Each idea is implemented and developed into a full paper at a cost of less than $15 per paper. To evaluate the generated papers, we design and validate an automated reviewer, which we show achieves near-human performance in evaluating paper scores. The AI Scientist can produce papers that exceed the acceptance threshold at a top machine learning conference as judged by our automated reviewer. This approach signifies the beginning of a new era in scientific discovery in machine learning: bringing the transformative benefits of AI agents to the entire research process of AI itself, and taking us closer to a world where endless affordable creativity and innovation can be unleashed on the world's most challenging problems. Our code is open-sourced at https://github.com/SakanaAI/AI-Scientist","external_url":"https://arxiv.org/abs/2408.06292","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-05-25T05:46:39.333419+00:00","pith_arxiv_id":"2408.06292","created_at":"2026-05-09T04:15:09.228737+00:00","updated_at":"2026-05-25T05:46:39.333419+00:00","title_quality_ok":true,"display_title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","render_title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery"},"hub":{"state":{"work_id":"56b6b58d-e73a-4317-896e-36ac5f84e957","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":127,"external_cited_by_count":null,"distinct_field_count":21,"first_pith_cited_at":"2024-01-27T03:53:25+00:00","last_pith_cited_at":"2026-05-22T15:27:09+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":"2026-06-03T16:56:12.313938+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":29},{"context_role":"baseline","n":2},{"context_role":"dataset","n":1},{"context_role":"method","n":1},{"context_role":"other","n":1}],"polarity_counts":[{"context_polarity":"background","n":26},{"context_polarity":"unclear","n":3},{"context_polarity":"baseline","n":2},{"context_polarity":"support","n":1},{"context_polarity":"use_dataset","n":1},{"context_polarity":"use_method","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","claims":[{"claim_text":"One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which ","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"capabilities across multiple modalities and disciplines. This section presents some prominent benchmarks designed to evaluate scientific reasoning, ranging from multimodal integration to specialized domain expertise, as summarized in Table 3. Table 3: Comparative Analysis of Scientific Reasoning Benchmarks Benchmark Scale Key Focus Domains Modality Difficulty ScienceQA [110] 21.2K questions Multimodal reason- ing Multidisciplinary Text+Visual Moderate MathVista [111] 6.1K questions Math+Visual r","claim_type":"dataset","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"or how the conclusions would change under a modified assumption. Current publications do not yet directly propose publishing the research agent itself, but adjacent work on end-to-end AI research systems, AI authorship and review, and publisher responses to AI in scientific communication suggests that scientific outputs are already moving toward more agent-mediated and interactive forms [ 17][18][19][20][21]. Such an agent could also adapt its explanation to different audiences. It might provide","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"39% accuracy on Deep Research and 9.31% IoU on Wide Research, while many other strong baselines fall be- low 5%. We publicly release the dataset, evaluation pipeline, and code at https://github.com/CherYou/AutoResearchBench. 1 Introduction The recent rise of LLM-basedAI scientists systemshas made autonomous scientific research a concrete target rather than a distant aspiration [1, 2, 3, 4, 5]. Across ideation, design, implementation, and experimentation, one capability is repeatedly indispensabl","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"variants [5] and synthesize post-training objectives [28]. These methods remain constrained by predefined search spaces or focus on optimizing isolated components. In contrast, our work explores a more open-ended setting, directly automating the entire LLM training lifecycle. AI for Data Construction.Recent studies extensively utilize LLMs for data synthesis [31], evolutionary refinement [48], and quality filtering [26]. To facilitate these tasks, dedicated frameworks [3, 24] have been 3 TREX: A","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Inspired by these developments, we ask whether the same agentic capabilities can be brought to bear onopen-ended discoveryin foundational theoretical computer science. We focus on the k-server conjecture, widely regarded as the \"holy grail of competitive analysis\" and a flagship open problem in online algorithms. Since its formulation by Manasse, McGeoch, and Sleator in 1990 [30], the conjecture has shaped the development of competitive analysis as a field [11, 10, 18, 25, 2, 3]. It asks whether","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"that NanoResearch delivers substantial gains over state-of-the-art AI research systems, and progressively refines itself to produce better research at lower cost over successive cycles. Code Dataset 1 Introduction LLM-powered multi-agent systems [1] have recently transformed end-to-end research automation from a long-standing aspiration [15, 31] into working reality. Systems such as The AI Scientist [21], AI Scientist-v2 [35], EvoScientist [22], and AI-Researcher [27] can now autonomously traver","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (27 contexts).","role_counts":[{"n":27,"context_role":"background"},{"n":2,"context_role":"baseline"},{"n":1,"context_role":"dataset"},{"n":1,"context_role":"other"}]},"error":null,"updated_at":"2026-05-20T11:21:55.930052+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"20311dbb-a2ac-4af1-8bec-95d456774bf4","orcid":null,"display_name":"Chris Lu"},{"id":"7bfde619-9d27-4d05-a003-6872d98225f5","orcid":null,"display_name":"Cong Lu"},{"id":"2e51a8c7-58b3-43a4-b6cb-27666cf5b4c2","orcid":null,"display_name":"Robert Tjarko Lange"},{"id":"e4bee8c4-a9c4-4d47-b211-e93d65b1a48a","orcid":null,"display_name":"Jakob Foerster"},{"id":"79c07cf8-e229-45d3-bbba-eb9e002878e3","orcid":null,"display_name":"Jeff Clune"},{"id":"ad438a60-54a3-4183-b69f-20c372ff34ee","orcid":null,"display_name":"David Ha"}]},"error":null,"updated_at":"2026-05-20T11:21:56.451993+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-14T09:07:55.598653+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"Towards an AI co-scientist","work_id":"485486b1-a1a2-4cde-bdda-768930c403e6","shared_citers":23},{"title":"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search","work_id":"fa04f346-ee20-4e9d-bf04-3ad3569a8ed1","shared_citers":20},{"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","shared_citers":12},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":12},{"title":"Ai-researcher: Autonomous scientific innovation","work_id":"3845f0f0-08d4-4650-b390-6bfdd269f79a","shared_citers":8},{"title":"Cycleresearcher: Improving automated research via automated review","work_id":"529c57f4-8402-4221-93f2-032f08d64085","shared_citers":6},{"title":"Deepscientist: Advancing frontier-pushing scientific findings progressively","work_id":"d68e01e1-6d49-4438-9cb6-114f102063a8","shared_citers":6},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":6},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":6},{"title":"OpenHands: An Open Platform for AI Software Developers as Generalist Agents","work_id":"f1762ea0-e382-4f38-a28c-adc643789859","shared_citers":6},{"title":"SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering","work_id":"01826cd9-a652-403c-a2ec-531da9fe2b6a","shared_citers":6},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":5},{"title":"GPT-4o System Card","work_id":"f37bf1c7-4964-4e56-9762-d20da8d9009f","shared_citers":5},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":5},{"title":"Researchagent: Iterative research idea generation over scientific literature with large language models","work_id":"41213a8f-51aa-4065-b3d5-2f154966db88","shared_citers":5},{"title":"SWE-bench: Can Language Models Resolve Real-World GitHub Issues?","work_id":"d0effe15-a689-441a-8e3f-ea35f1c4e4b1","shared_citers":5},{"title":"Voyager: An Open-Ended Embodied Agent with Large Language Models","work_id":"ffe0d207-86cf-4742-a100-e988ac8b9676","shared_citers":5},{"title":"$\\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains","work_id":"6a8d8dc4-0cc0-4052-8109-abbcdcd4a962","shared_citers":4},{"title":"2310.03302 , archivePrefix =","work_id":"5655b20a-fbb7-4a39-8605-d6e1d689895a","shared_citers":4},{"title":"Agent laboratory: Using llm agents as research assistants, 2025.URL https://arxiv","work_id":"cb6e0bd7-2f2d-4052-875a-985ba5600340","shared_citers":4},{"title":"Aide: Ai-driven exploration in the space of code","work_id":"22aa3d2a-9edd-44c4-b8b8-1442ea805e01","shared_citers":4},{"title":"Can llms generate novel research ideas? a large-scale human study with 100+ nlp researchers","work_id":"95c4070f-48fa-44f6-bed8-a4b874c54eac","shared_citers":4},{"title":"From automation to autonomy: A survey on large language models in scientific discovery","work_id":"631dea7e-c36b-4852-a827-bf6b15ebf453","shared_citers":4},{"title":"Georgiev, J","work_id":"b3882729-74ba-4d80-9240-bfe03c45c33b","shared_citers":4}],"time_series":[{"n":1,"year":2025},{"n":68,"year":2026}],"dependency_candidates":[]},"error":null,"updated_at":"2026-05-14T09:18:07.906591+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":"2026-05-14T09:08:04.096149+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","claims":[{"claim_text":"One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which ","claim_type":"abstract","evidence_strength":"source_metadata"},{"claim_text":"capabilities across multiple modalities and disciplines. This section presents some prominent benchmarks designed to evaluate scientific reasoning, ranging from multimodal integration to specialized domain expertise, as summarized in Table 3. Table 3: Comparative Analysis of Scientific Reasoning Benchmarks Benchmark Scale Key Focus Domains Modality Difficulty ScienceQA [110] 21.2K questions Multimodal reason- ing Multidisciplinary Text+Visual Moderate MathVista [111] 6.1K questions Math+Visual r","claim_type":"dataset","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"or how the conclusions would change under a modified assumption. Current publications do not yet directly propose publishing the research agent itself, but adjacent work on end-to-end AI research systems, AI authorship and review, and publisher responses to AI in scientific communication suggests that scientific outputs are already moving toward more agent-mediated and interactive forms [ 17][18][19][20][21]. Such an agent could also adapt its explanation to different audiences. It might provide","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"39% accuracy on Deep Research and 9.31% IoU on Wide Research, while many other strong baselines fall be- low 5%. We publicly release the dataset, evaluation pipeline, and code at https://github.com/CherYou/AutoResearchBench. 1 Introduction The recent rise of LLM-basedAI scientists systemshas made autonomous scientific research a concrete target rather than a distant aspiration [1, 2, 3, 4, 5]. Across ideation, design, implementation, and experimentation, one capability is repeatedly indispensabl","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"variants [5] and synthesize post-training objectives [28]. These methods remain constrained by predefined search spaces or focus on optimizing isolated components. In contrast, our work explores a more open-ended setting, directly automating the entire LLM training lifecycle. AI for Data Construction.Recent studies extensively utilize LLMs for data synthesis [31], evolutionary refinement [48], and quality filtering [26]. To facilitate these tasks, dedicated frameworks [3, 24] have been 3 TREX: A","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"Inspired by these developments, we ask whether the same agentic capabilities can be brought to bear onopen-ended discoveryin foundational theoretical computer science. We focus on the k-server conjecture, widely regarded as the \"holy grail of competitive analysis\" and a flagship open problem in online algorithms. Since its formulation by Manasse, McGeoch, and Sleator in 1990 [30], the conjecture has shaped the development of competitive analysis as a field [11, 10, 18, 25, 2, 3]. It asks whether","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"},{"claim_text":"that NanoResearch delivers substantial gains over state-of-the-art AI research systems, and progressively refines itself to produce better research at lower cost over successive cycles. Code Dataset 1 Introduction LLM-powered multi-agent systems [1] have recently transformed end-to-end research automation from a long-standing aspiration [15, 31] into working reality. Systems such as The AI Scientist [21], AI Scientist-v2 [35], EvoScientist [22], and AI-Researcher [27] can now autonomously traver","claim_type":"background","confidence":0.9,"evidence_strength":"citation_context"}],"why_cited":"Pith tracks The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery because it crossed a citation-hub threshold. Current citing contexts most often use it as background evidence (27 contexts).","role_counts":[{"n":27,"context_role":"background"},{"n":2,"context_role":"baseline"},{"n":1,"context_role":"dataset"},{"n":1,"context_role":"other"}]},"error":null,"updated_at":"2026-05-20T11:21:55.606978+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","claims":[{"claim_text":"One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which ","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-14T09:18:07.909548+00:00"}},"summary":{"title":"The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery","claims":[{"claim_text":"One of the grand challenges of artificial general intelligence is developing agents capable of conducting scientific research and discovering new knowledge. While frontier models have already been used as aides to human scientists, e.g. for brainstorming ideas, writing code, or prediction tasks, they still conduct only a small part of the scientific process. This paper presents the first comprehensive framework for fully automatic scientific discovery, enabling frontier large language models to perform research independently and communicate their findings. We introduce The AI Scientist, which ","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"Towards an AI co-scientist","work_id":"485486b1-a1a2-4cde-bdda-768930c403e6","shared_citers":23},{"title":"The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search","work_id":"fa04f346-ee20-4e9d-bf04-3ad3569a8ed1","shared_citers":20},{"title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","work_id":"76a0f850-d490-4e4f-ab98-8d25df82cd23","shared_citers":12},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":12},{"title":"Ai-researcher: Autonomous scientific innovation","work_id":"3845f0f0-08d4-4650-b390-6bfdd269f79a","shared_citers":8},{"title":"Cycleresearcher: Improving automated research via automated review","work_id":"529c57f4-8402-4221-93f2-032f08d64085","shared_citers":6},{"title":"Deepscientist: Advancing frontier-pushing scientific findings progressively","work_id":"d68e01e1-6d49-4438-9cb6-114f102063a8","shared_citers":6},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":6},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":6},{"title":"OpenHands: An Open Platform for AI Software Developers as Generalist Agents","work_id":"f1762ea0-e382-4f38-a28c-adc643789859","shared_citers":6},{"title":"SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering","work_id":"01826cd9-a652-403c-a2ec-531da9fe2b6a","shared_citers":6},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":5},{"title":"GPT-4o System Card","work_id":"f37bf1c7-4964-4e56-9762-d20da8d9009f","shared_citers":5},{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":5},{"title":"Researchagent: Iterative research idea generation over scientific literature with large language models","work_id":"41213a8f-51aa-4065-b3d5-2f154966db88","shared_citers":5},{"title":"SWE-bench: Can Language Models Resolve Real-World GitHub Issues?","work_id":"d0effe15-a689-441a-8e3f-ea35f1c4e4b1","shared_citers":5},{"title":"Voyager: An Open-Ended Embodied Agent with Large Language Models","work_id":"ffe0d207-86cf-4742-a100-e988ac8b9676","shared_citers":5},{"title":"$\\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains","work_id":"6a8d8dc4-0cc0-4052-8109-abbcdcd4a962","shared_citers":4},{"title":"2310.03302 , archivePrefix =","work_id":"5655b20a-fbb7-4a39-8605-d6e1d689895a","shared_citers":4},{"title":"Agent laboratory: Using llm agents as research assistants, 2025.URL https://arxiv","work_id":"cb6e0bd7-2f2d-4052-875a-985ba5600340","shared_citers":4},{"title":"Aide: Ai-driven exploration in the space of code","work_id":"22aa3d2a-9edd-44c4-b8b8-1442ea805e01","shared_citers":4},{"title":"Can llms generate novel research ideas? a large-scale human study with 100+ nlp researchers","work_id":"95c4070f-48fa-44f6-bed8-a4b874c54eac","shared_citers":4},{"title":"From automation to autonomy: A survey on large language models in scientific discovery","work_id":"631dea7e-c36b-4852-a827-bf6b15ebf453","shared_citers":4},{"title":"Georgiev, J","work_id":"b3882729-74ba-4d80-9240-bfe03c45c33b","shared_citers":4}],"time_series":[{"n":1,"year":2025},{"n":68,"year":2026}],"dependency_candidates":[]},"authors":[{"id":"20311dbb-a2ac-4af1-8bec-95d456774bf4","orcid":null,"display_name":"Chris Lu","source":"manual","import_confidence":0.72},{"id":"7bfde619-9d27-4d05-a003-6872d98225f5","orcid":null,"display_name":"Cong Lu","source":"manual","import_confidence":0.72},{"id":"ad438a60-54a3-4183-b69f-20c372ff34ee","orcid":null,"display_name":"David Ha","source":"manual","import_confidence":0.72},{"id":"e4bee8c4-a9c4-4d47-b211-e93d65b1a48a","orcid":null,"display_name":"Jakob Foerster","source":"manual","import_confidence":0.72},{"id":"79c07cf8-e229-45d3-bbba-eb9e002878e3","orcid":null,"display_name":"Jeff Clune","source":"manual","import_confidence":0.72},{"id":"2e51a8c7-58b3-43a4-b6cb-27666cf5b4c2","orcid":null,"display_name":"Robert Tjarko Lange","source":"manual","import_confidence":0.72}]}}