{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6F4BAM4ZD3ILSSGE7OACJ3RR2W","short_pith_number":"pith:6F4BAM4Z","schema_version":"1.0","canonical_sha256":"f1781033991ed0b948c4fb8024ee31d58dd7d28dd9ffc566b316b18819c3d749","source":{"kind":"arxiv","id":"2606.26300","version":1},"attestation_state":"computed","paper":{"title":"The Verification Horizon: No Silver Bullet for Coding Agent Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Binghai Wang, Chenlong Zhang, Dayiheng Liu, Jiajun Zhang, Jiawei Chen, Mouxiang Chen, Rongyao Fang, Siyuan Zhang, Xuwu Wang, Yuheng Jing, Zeyao Ma, Zeyu Cui","submitted_at":"2026-06-24T18:45:03Z","abstract_excerpt":"A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabilities and engineering harnesses grow more sophisticated, generating complex candidate solutions is no longer difficult -- reliably verifying them has become the harder problem. Every verifier we can build is only a proxy for human intent, never the intent itself. This makes verification subject to a twofold difficulty: first, intent is underspecified by nature, making it inherently hard to faithfu"},"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":"2606.26300","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T18:45:03Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"36ba8790b779e6e74c27a221fea46f2087507637cc230094dc5850d0af7d0c26","abstract_canon_sha256":"18e11e67e3fe03a6ad27f44ca7010e14f37e55fd54f8aec295ae977e2308ae97"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T00:15:35.969359Z","signature_b64":"yvMdXS+gMu5/xyrGz1bjBrdX2pDBlP9DVAAbUYNcJtIfJjmsNs+VhS2yiGi85MzYp6ZOKWHb+jXnNAredQSXBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1781033991ed0b948c4fb8024ee31d58dd7d28dd9ffc566b316b18819c3d749","last_reissued_at":"2026-06-26T00:15:35.968887Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T00:15:35.968887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Verification Horizon: No Silver Bullet for Coding Agent Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Binghai Wang, Chenlong Zhang, Dayiheng Liu, Jiajun Zhang, Jiawei Chen, Mouxiang Chen, Rongyao Fang, Siyuan Zhang, Xuwu Wang, Yuheng Jing, Zeyao Ma, Zeyu Cui","submitted_at":"2026-06-24T18:45:03Z","abstract_excerpt":"A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabilities and engineering harnesses grow more sophisticated, generating complex candidate solutions is no longer difficult -- reliably verifying them has become the harder problem. Every verifier we can build is only a proxy for human intent, never the intent itself. This makes verification subject to a twofold difficulty: first, intent is underspecified by nature, making it inherently hard to faithfu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26300","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/2606.26300/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":"2606.26300","created_at":"2026-06-26T00:15:35.968963+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26300v1","created_at":"2026-06-26T00:15:35.968963+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26300","created_at":"2026-06-26T00:15:35.968963+00:00"},{"alias_kind":"pith_short_12","alias_value":"6F4BAM4ZD3IL","created_at":"2026-06-26T00:15:35.968963+00:00"},{"alias_kind":"pith_short_16","alias_value":"6F4BAM4ZD3ILSSGE","created_at":"2026-06-26T00:15:35.968963+00:00"},{"alias_kind":"pith_short_8","alias_value":"6F4BAM4Z","created_at":"2026-06-26T00:15:35.968963+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/6F4BAM4ZD3ILSSGE7OACJ3RR2W","json":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W.json","graph_json":"https://pith.science/api/pith-number/6F4BAM4ZD3ILSSGE7OACJ3RR2W/graph.json","events_json":"https://pith.science/api/pith-number/6F4BAM4ZD3ILSSGE7OACJ3RR2W/events.json","paper":"https://pith.science/paper/6F4BAM4Z"},"agent_actions":{"view_html":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W","download_json":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W.json","view_paper":"https://pith.science/paper/6F4BAM4Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26300&json=true","fetch_graph":"https://pith.science/api/pith-number/6F4BAM4ZD3ILSSGE7OACJ3RR2W/graph.json","fetch_events":"https://pith.science/api/pith-number/6F4BAM4ZD3ILSSGE7OACJ3RR2W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W/action/storage_attestation","attest_author":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W/action/author_attestation","sign_citation":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W/action/citation_signature","submit_replication":"https://pith.science/pith/6F4BAM4ZD3ILSSGE7OACJ3RR2W/action/replication_record"}},"created_at":"2026-06-26T00:15:35.968963+00:00","updated_at":"2026-06-26T00:15:35.968963+00:00"}