{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A5AHTY4ZYJ5QMQJZJV366PKD55","short_pith_number":"pith:A5AHTY4Z","schema_version":"1.0","canonical_sha256":"074079e399c27b0641394d77ef3d43ef43a4d6f7f8f1d6fbd3fc639c1e19c202","source":{"kind":"arxiv","id":"2506.10055","version":2},"attestation_state":"computed","paper":{"title":"TaskCraft: Automated Generation of Agentic Tasks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changwang Zhang, Dingfeng Shi, Fangchen Dong, Ge Zhang, Hongxuan Lu, Jiaheng Liu, Jian Yang, Jingyi Cao, Jun Wang, King Zhu, Minghao Liu, Qianben Chen, Tianrui Qin, Wangchunshu Zhou, Weichen Sun, Weizhen Li, Yuchen Eleanor Jiang","submitted_at":"2025-06-11T17:58:14Z","abstract_excerpt":"Agentic tasks, which require multi-step problem solving with autonomy, tool use, and adaptive reasoning, are becoming increasingly central to the advancement of NLP and AI. However, existing instruction data lacks tool interaction, and current agentic benchmarks rely on costly human annotation, limiting their scalability. We introduce \\textsc{TaskCraft}, an automated workflow for generating difficulty-scalable, multi-tool, and verifiable agentic tasks with execution trajectories. TaskCraft expands atomic tasks using depth-based and width-based extensions to create structurally and hierarchical"},"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":"2506.10055","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T17:58:14Z","cross_cats_sorted":[],"title_canon_sha256":"9adba72bd12329b18dcbc6948f2dc570a54aa9fa6d753291332ffb851e0184c3","abstract_canon_sha256":"ea000e4ef3646e48d0f6c570397e164be476566a42f2d54f2c06b4a1de33992c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:41.896132Z","signature_b64":"AwnHZyJHd46Wl0hz6ibu47dYCRLx4pxFIfeFvxN/jjyr3uyRp4mnRea2Pf5yy3rBpWKLukZdoWztvvvobLZNAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"074079e399c27b0641394d77ef3d43ef43a4d6f7f8f1d6fbd3fc639c1e19c202","last_reissued_at":"2026-07-05T11:22:41.895653Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:41.895653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TaskCraft: Automated Generation of Agentic Tasks","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Changwang Zhang, Dingfeng Shi, Fangchen Dong, Ge Zhang, Hongxuan Lu, Jiaheng Liu, Jian Yang, Jingyi Cao, Jun Wang, King Zhu, Minghao Liu, Qianben Chen, Tianrui Qin, Wangchunshu Zhou, Weichen Sun, Weizhen Li, Yuchen Eleanor Jiang","submitted_at":"2025-06-11T17:58:14Z","abstract_excerpt":"Agentic tasks, which require multi-step problem solving with autonomy, tool use, and adaptive reasoning, are becoming increasingly central to the advancement of NLP and AI. However, existing instruction data lacks tool interaction, and current agentic benchmarks rely on costly human annotation, limiting their scalability. We introduce \\textsc{TaskCraft}, an automated workflow for generating difficulty-scalable, multi-tool, and verifiable agentic tasks with execution trajectories. TaskCraft expands atomic tasks using depth-based and width-based extensions to create structurally and hierarchical"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10055","kind":"arxiv","version":2},"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/2506.10055/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":"2506.10055","created_at":"2026-07-05T11:22:41.895711+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.10055v2","created_at":"2026-07-05T11:22:41.895711+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10055","created_at":"2026-07-05T11:22:41.895711+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5AHTY4ZYJ5Q","created_at":"2026-07-05T11:22:41.895711+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5AHTY4ZYJ5QMQJZ","created_at":"2026-07-05T11:22:41.895711+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5AHTY4Z","created_at":"2026-07-05T11:22:41.895711+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.17931","citing_title":"LiteResearcher: A Scalable Agentic RL Training Framework for Deep Research Agent","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2606.11520","citing_title":"ISE: An Execution-Grounded Recipe for Multi-Turn OS-Agent Trajectories","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07074","citing_title":"SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12191","citing_title":"Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application","ref_index":177,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22138","citing_title":"Efficient Agentic Reasoning Through Self-Regulated Simulative Planning","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20876","citing_title":"Terminal-World: Scaling Terminal-Agent Environments via Agent Skills","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18181","citing_title":"Scalable Environments Drive Generalizable Agents","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2511.11793","citing_title":"MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03044","citing_title":"JoyAI-LLM Flash: Advancing Mid-Scale LLMs with Token Efficiency","ref_index":67,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55","json":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55.json","graph_json":"https://pith.science/api/pith-number/A5AHTY4ZYJ5QMQJZJV366PKD55/graph.json","events_json":"https://pith.science/api/pith-number/A5AHTY4ZYJ5QMQJZJV366PKD55/events.json","paper":"https://pith.science/paper/A5AHTY4Z"},"agent_actions":{"view_html":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55","download_json":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55.json","view_paper":"https://pith.science/paper/A5AHTY4Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.10055&json=true","fetch_graph":"https://pith.science/api/pith-number/A5AHTY4ZYJ5QMQJZJV366PKD55/graph.json","fetch_events":"https://pith.science/api/pith-number/A5AHTY4ZYJ5QMQJZJV366PKD55/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55/action/storage_attestation","attest_author":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55/action/author_attestation","sign_citation":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55/action/citation_signature","submit_replication":"https://pith.science/pith/A5AHTY4ZYJ5QMQJZJV366PKD55/action/replication_record"}},"created_at":"2026-07-05T11:22:41.895711+00:00","updated_at":"2026-07-05T11:22:41.895711+00:00"}