{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CAUJSOG2AA4VEAH2SMEEJYD5LD","short_pith_number":"pith:CAUJSOG2","schema_version":"1.0","canonical_sha256":"10289938da00395200fa930844e07d58fbf717ec4162cd81d6ac79ab8db1383d","source":{"kind":"arxiv","id":"2607.17082","version":1},"attestation_state":"computed","paper":{"title":"Otap:Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Babak Barazandeh, George Michailidis, Subhabrata Majumdar","submitted_at":"2026-07-19T05:23:17Z","abstract_excerpt":"Large language model agents solve tasks by generating trajectories that interleave planning, tool calls, and intermediate results. Current evaluation metrics reduce such a trajectory to a binary success flag or compare it against a reference by exact matching. A success flag cannot distinguish a sound solution from one that succeeds by luck, and says nothing about why a failed run went wrong. Exact matching penalizes plans that are valid but reordered or decomposed differently from the reference. We reframe trajectory evaluation as a distance between the agent's execution graph and a set of va"},"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":"2607.17082","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-19T05:23:17Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"526a67559a76f82a407cd8826f4890dc7af6f48c6ca3833391f3821b4aa43e35","abstract_canon_sha256":"062e3401bba6aabb12e355ca27afff0b5523242dfde4ae4ae229d214fc35b57a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:13.959063Z","signature_b64":"mt8uNDEBKTRCtd/Oq8expZ7zd17QYPUZyszZGsD0HC2NuJ8jJh0f2UeyIu6Npq8//VhRiPIZp+WY0qAj1Ty3CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"10289938da00395200fa930844e07d58fbf717ec4162cd81d6ac79ab8db1383d","last_reissued_at":"2026-07-21T01:21:13.957986Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:13.957986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Otap:Structure-Aware Optimal Transport for Evaluating Planning and Execution in Agent Trajectories","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.AI","authors_text":"Babak Barazandeh, George Michailidis, Subhabrata Majumdar","submitted_at":"2026-07-19T05:23:17Z","abstract_excerpt":"Large language model agents solve tasks by generating trajectories that interleave planning, tool calls, and intermediate results. Current evaluation metrics reduce such a trajectory to a binary success flag or compare it against a reference by exact matching. A success flag cannot distinguish a sound solution from one that succeeds by luck, and says nothing about why a failed run went wrong. Exact matching penalizes plans that are valid but reordered or decomposed differently from the reference. We reframe trajectory evaluation as a distance between the agent's execution graph and a set of va"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17082","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/2607.17082/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":"2607.17082","created_at":"2026-07-21T01:21:13.958597+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.17082v1","created_at":"2026-07-21T01:21:13.958597+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17082","created_at":"2026-07-21T01:21:13.958597+00:00"},{"alias_kind":"pith_short_12","alias_value":"CAUJSOG2AA4V","created_at":"2026-07-21T01:21:13.958597+00:00"},{"alias_kind":"pith_short_16","alias_value":"CAUJSOG2AA4VEAH2","created_at":"2026-07-21T01:21:13.958597+00:00"},{"alias_kind":"pith_short_8","alias_value":"CAUJSOG2","created_at":"2026-07-21T01:21:13.958597+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/CAUJSOG2AA4VEAH2SMEEJYD5LD","json":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD.json","graph_json":"https://pith.science/api/pith-number/CAUJSOG2AA4VEAH2SMEEJYD5LD/graph.json","events_json":"https://pith.science/api/pith-number/CAUJSOG2AA4VEAH2SMEEJYD5LD/events.json","paper":"https://pith.science/paper/CAUJSOG2"},"agent_actions":{"view_html":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD","download_json":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD.json","view_paper":"https://pith.science/paper/CAUJSOG2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.17082&json=true","fetch_graph":"https://pith.science/api/pith-number/CAUJSOG2AA4VEAH2SMEEJYD5LD/graph.json","fetch_events":"https://pith.science/api/pith-number/CAUJSOG2AA4VEAH2SMEEJYD5LD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD/action/storage_attestation","attest_author":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD/action/author_attestation","sign_citation":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD/action/citation_signature","submit_replication":"https://pith.science/pith/CAUJSOG2AA4VEAH2SMEEJYD5LD/action/replication_record"}},"created_at":"2026-07-21T01:21:13.958597+00:00","updated_at":"2026-07-21T01:21:13.958597+00:00"}