{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3VP3X4RFTKV7JVVN3UKY7ILLEP","short_pith_number":"pith:3VP3X4RF","schema_version":"1.0","canonical_sha256":"dd5fbbf2259aabf4d6addd158fa16b23cf659f9eef6438488b5079891b340e9c","source":{"kind":"arxiv","id":"2605.23574","version":1},"attestation_state":"computed","paper":{"title":"Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.LG","authors_text":"Liyou Gao, Shengchao Qin, Wensheng Tang, Yuandao Cai, Yuzhang Zhu","submitted_at":"2026-05-22T12:44:01Z","abstract_excerpt":"Long-horizon language agents can make many plausible local tool calls yet fail to persist until a requested count is actually complete. We study this gap as Quantitative Goal Persistence (QGP): whether an agent keeps working until an external verifier confirms enough distinct valid items. PushBench turns this into a benchmark for repository-artifact collection and verifier-backed work units, so repeated work, duplicate submissions, false completion, and progress drift are measured directly rather than hidden behind a final success flag. In matched controller comparisons, a state-tracking retri"},"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":"2605.23574","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-22T12:44:01Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"3dc0481e1d6e851bd87fea1cca3120d596ba57893b1f9957ef7ba18888864f6f","abstract_canon_sha256":"0424ca42d0185da41e0493d4a4cfe43fe6b01731344d2ca48469279e28734877"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:02:20.284086Z","signature_b64":"OzInCFv6YeHbxi0EQcFykh6z5DiWNW7ezVC9hk/KhCI1D6tyht1uJ2SOrJHPAzSfsufPYTJo1qlyDOMHrATSBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dd5fbbf2259aabf4d6addd158fa16b23cf659f9eef6438488b5079891b340e9c","last_reissued_at":"2026-05-25T02:02:20.283285Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:02:20.283285Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Push Your Agent: Measuring and Enforcing Quantitative Goal Persistence in Long-Horizon LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.LG","authors_text":"Liyou Gao, Shengchao Qin, Wensheng Tang, Yuandao Cai, Yuzhang Zhu","submitted_at":"2026-05-22T12:44:01Z","abstract_excerpt":"Long-horizon language agents can make many plausible local tool calls yet fail to persist until a requested count is actually complete. We study this gap as Quantitative Goal Persistence (QGP): whether an agent keeps working until an external verifier confirms enough distinct valid items. PushBench turns this into a benchmark for repository-artifact collection and verifier-backed work units, so repeated work, duplicate submissions, false completion, and progress drift are measured directly rather than hidden behind a final success flag. In matched controller comparisons, a state-tracking retri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23574","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/2605.23574/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":"2605.23574","created_at":"2026-05-25T02:02:20.283429+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.23574v1","created_at":"2026-05-25T02:02:20.283429+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23574","created_at":"2026-05-25T02:02:20.283429+00:00"},{"alias_kind":"pith_short_12","alias_value":"3VP3X4RFTKV7","created_at":"2026-05-25T02:02:20.283429+00:00"},{"alias_kind":"pith_short_16","alias_value":"3VP3X4RFTKV7JVVN","created_at":"2026-05-25T02:02:20.283429+00:00"},{"alias_kind":"pith_short_8","alias_value":"3VP3X4RF","created_at":"2026-05-25T02:02:20.283429+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/3VP3X4RFTKV7JVVN3UKY7ILLEP","json":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP.json","graph_json":"https://pith.science/api/pith-number/3VP3X4RFTKV7JVVN3UKY7ILLEP/graph.json","events_json":"https://pith.science/api/pith-number/3VP3X4RFTKV7JVVN3UKY7ILLEP/events.json","paper":"https://pith.science/paper/3VP3X4RF"},"agent_actions":{"view_html":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP","download_json":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP.json","view_paper":"https://pith.science/paper/3VP3X4RF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.23574&json=true","fetch_graph":"https://pith.science/api/pith-number/3VP3X4RFTKV7JVVN3UKY7ILLEP/graph.json","fetch_events":"https://pith.science/api/pith-number/3VP3X4RFTKV7JVVN3UKY7ILLEP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP/action/storage_attestation","attest_author":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP/action/author_attestation","sign_citation":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP/action/citation_signature","submit_replication":"https://pith.science/pith/3VP3X4RFTKV7JVVN3UKY7ILLEP/action/replication_record"}},"created_at":"2026-05-25T02:02:20.283429+00:00","updated_at":"2026-05-25T02:02:20.283429+00:00"}