{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:J5YDACLRAHSKRFY25CYPXXXU4E","short_pith_number":"pith:J5YDACLR","schema_version":"1.0","canonical_sha256":"4f7030097101e4a8971ae8b0fbdef4e1315d5343dda1b9f27ae3ccdc87cb2106","source":{"kind":"arxiv","id":"2508.08053","version":1},"attestation_state":"computed","paper":{"title":"AdaptFlow: Adaptive Workflow Optimization via Meta-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bowen Jiang, Dongmei Zhang, Fangkai Yang, Fengshuo Bai, Haoxiang Gao, Lingrui Mei, Lu Wang, Pu Zhao, Qingwei Lin, Runchuan Zhu, Saravan Rajmohan","submitted_at":"2025-08-11T14:52:59Z","abstract_excerpt":"Recent advances in large language models (LLMs) have sparked growing interest in agentic workflows, which are structured sequences of LLM invocations intended to solve complex tasks. However, existing approaches often rely on static templates or manually designed workflows, which limit adaptability to diverse tasks and hinder scalability. We propose AdaptFlow, a natural language-based meta-learning framework inspired by model-agnostic meta-learning (MAML). AdaptFlow learns a generalizable workflow initialization that enables rapid subtask-level adaptation. It employs a bi-level optimization sc"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2508.08053","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-11T14:52:59Z","cross_cats_sorted":[],"title_canon_sha256":"b3a6c13d8166cffa10717562ceff753b426470e029af1cdc10522f4686a11e9e","abstract_canon_sha256":"efdd81a1261f5ee6b499f0bc507d57bd8fb07e630bb3f01e9d768f57a182f850"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:07.174570Z","signature_b64":"rELr9X2NtM42DZ+KfiNzsjFVYx+XP0tQ2UjjFf8sQxUqZfwhFRLcFIOE1kbKYNYGSuW6NcCEOJr8Q+dxs4bTDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f7030097101e4a8971ae8b0fbdef4e1315d5343dda1b9f27ae3ccdc87cb2106","last_reissued_at":"2026-07-05T11:52:07.173949Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:07.173949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AdaptFlow: Adaptive Workflow Optimization via Meta-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bowen Jiang, Dongmei Zhang, Fangkai Yang, Fengshuo Bai, Haoxiang Gao, Lingrui Mei, Lu Wang, Pu Zhao, Qingwei Lin, Runchuan Zhu, Saravan Rajmohan","submitted_at":"2025-08-11T14:52:59Z","abstract_excerpt":"Recent advances in large language models (LLMs) have sparked growing interest in agentic workflows, which are structured sequences of LLM invocations intended to solve complex tasks. However, existing approaches often rely on static templates or manually designed workflows, which limit adaptability to diverse tasks and hinder scalability. We propose AdaptFlow, a natural language-based meta-learning framework inspired by model-agnostic meta-learning (MAML). AdaptFlow learns a generalizable workflow initialization that enables rapid subtask-level adaptation. It employs a bi-level optimization sc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.08053","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.08053/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2508.08053","created_at":"2026-07-05T11:52:07.174011+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.08053v1","created_at":"2026-07-05T11:52:07.174011+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.08053","created_at":"2026-07-05T11:52:07.174011+00:00"},{"alias_kind":"pith_short_12","alias_value":"J5YDACLRAHSK","created_at":"2026-07-05T11:52:07.174011+00:00"},{"alias_kind":"pith_short_16","alias_value":"J5YDACLRAHSKRFY2","created_at":"2026-07-05T11:52:07.174011+00:00"},{"alias_kind":"pith_short_8","alias_value":"J5YDACLR","created_at":"2026-07-05T11:52:07.174011+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22385","citing_title":"MetaPS: Adaptive Programmatic Strategy Selection for Market Agents","ref_index":71,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E","json":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E.json","graph_json":"https://pith.science/api/pith-number/J5YDACLRAHSKRFY25CYPXXXU4E/graph.json","events_json":"https://pith.science/api/pith-number/J5YDACLRAHSKRFY25CYPXXXU4E/events.json","paper":"https://pith.science/paper/J5YDACLR"},"agent_actions":{"view_html":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E","download_json":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E.json","view_paper":"https://pith.science/paper/J5YDACLR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.08053&json=true","fetch_graph":"https://pith.science/api/pith-number/J5YDACLRAHSKRFY25CYPXXXU4E/graph.json","fetch_events":"https://pith.science/api/pith-number/J5YDACLRAHSKRFY25CYPXXXU4E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E/action/storage_attestation","attest_author":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E/action/author_attestation","sign_citation":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E/action/citation_signature","submit_replication":"https://pith.science/pith/J5YDACLRAHSKRFY25CYPXXXU4E/action/replication_record"}},"created_at":"2026-07-05T11:52:07.174011+00:00","updated_at":"2026-07-05T11:52:07.174011+00:00"}