{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IYWWMQQCCIQFLQVD2O5SPOGKIK","short_pith_number":"pith:IYWWMQQC","schema_version":"1.0","canonical_sha256":"462d664202122055c2a3d3bb27b8ca42a0c115ac883f4481e7cad996baa1720d","source":{"kind":"arxiv","id":"2608.12282","version":1},"attestation_state":"computed","paper":{"title":"VAKRA: Evaluating Multi-Hop Reasoning Across APIs and Retrieval Under Tool-Use Policies","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Abdulhamid Adebayo, Abhinav Jain, Ankita Rajaram Naik, Anupama Murthi, Benjamin Elder, Danish Contractor, Praveen Venkateswaran, Raavi Gupta, Siyu Huo","submitted_at":"2026-08-12T17:27:27Z","abstract_excerpt":"Agents deployed in enterprise settings must reason across structured APIs and document collections, yet existing benchmarks evaluate these capabilities in isolation. We introduce VAKRA (e\\textbf{V}aluating \\textbf{A}PI and \\textbf{K}nowledge \\textbf{R}etrieval \\textbf{A}gents), a benchmark of over $8{,}000$ executable APIs across $62$ domains with tasks spanning three settings of increasing difficulty: diverse API interaction styles, multi-hop reasoning over structured APIs, and multi-source reasoning with natural-language tool-use policy constraints. Correctness is verified by re-executing pr"},"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":"2608.12282","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-12T17:27:27Z","cross_cats_sorted":[],"title_canon_sha256":"211f87e2d4b14e8f4505829290e5786cdedbf3765988d88f91931e1a1b001e85","abstract_canon_sha256":"a9753cd943ac800cbe20ea85b03457fc940473cd0df2618883fe95e853a890b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T01:30:42.735918Z","signature_b64":"oUplQk7sOLrXaTeFSkgqTKnrxAY9Iegv/9iRR0i+OJTsiF7JQCJTqRrmwiokSwGPmWPPg309rbc4g4k+acXPAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"462d664202122055c2a3d3bb27b8ca42a0c115ac883f4481e7cad996baa1720d","last_reissued_at":"2026-08-13T01:30:42.733560Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T01:30:42.733560Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VAKRA: Evaluating Multi-Hop Reasoning Across APIs and Retrieval Under Tool-Use Policies","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Abdulhamid Adebayo, Abhinav Jain, Ankita Rajaram Naik, Anupama Murthi, Benjamin Elder, Danish Contractor, Praveen Venkateswaran, Raavi Gupta, Siyu Huo","submitted_at":"2026-08-12T17:27:27Z","abstract_excerpt":"Agents deployed in enterprise settings must reason across structured APIs and document collections, yet existing benchmarks evaluate these capabilities in isolation. We introduce VAKRA (e\\textbf{V}aluating \\textbf{A}PI and \\textbf{K}nowledge \\textbf{R}etrieval \\textbf{A}gents), a benchmark of over $8{,}000$ executable APIs across $62$ domains with tasks spanning three settings of increasing difficulty: diverse API interaction styles, multi-hop reasoning over structured APIs, and multi-source reasoning with natural-language tool-use policy constraints. Correctness is verified by re-executing pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.12282","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/2608.12282/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":"2608.12282","created_at":"2026-08-13T01:30:42.734804+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.12282v1","created_at":"2026-08-13T01:30:42.734804+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.12282","created_at":"2026-08-13T01:30:42.734804+00:00"},{"alias_kind":"pith_short_12","alias_value":"IYWWMQQCCIQF","created_at":"2026-08-13T01:30:42.734804+00:00"},{"alias_kind":"pith_short_16","alias_value":"IYWWMQQCCIQFLQVD","created_at":"2026-08-13T01:30:42.734804+00:00"},{"alias_kind":"pith_short_8","alias_value":"IYWWMQQC","created_at":"2026-08-13T01:30:42.734804+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK","json":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK.json","graph_json":"https://pith.science/api/pith-number/IYWWMQQCCIQFLQVD2O5SPOGKIK/graph.json","events_json":"https://pith.science/api/pith-number/IYWWMQQCCIQFLQVD2O5SPOGKIK/events.json","paper":"https://pith.science/paper/IYWWMQQC"},"agent_actions":{"view_html":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK","download_json":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK.json","view_paper":"https://pith.science/paper/IYWWMQQC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.12282&json=true","fetch_graph":"https://pith.science/api/pith-number/IYWWMQQCCIQFLQVD2O5SPOGKIK/graph.json","fetch_events":"https://pith.science/api/pith-number/IYWWMQQCCIQFLQVD2O5SPOGKIK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK/action/storage_attestation","attest_author":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK/action/author_attestation","sign_citation":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK/action/citation_signature","submit_replication":"https://pith.science/pith/IYWWMQQCCIQFLQVD2O5SPOGKIK/action/replication_record"}},"created_at":"2026-08-13T01:30:42.734804+00:00","updated_at":"2026-08-13T01:30:42.734804+00:00"}