{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6XEYCQAURY54IF4FFNLELAYZNL","short_pith_number":"pith:6XEYCQAU","schema_version":"1.0","canonical_sha256":"f5c98140148e3bc417852b564583196af49dda892392db0586eeb0be3bbac13c","source":{"kind":"arxiv","id":"2606.14948","version":2},"attestation_state":"computed","paper":{"title":"Beyond Correctness: Enhancing Architectural Reasoning in Code LLMs via Scalable Labeling with Agentic Judgment","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Ahmed E. Hassan, Arthur Leung, Benjamin Rombaut, Boyuan Chen, Dayi Lin, Jiahuei Lin (Justina), Kirill Vasilevski, Milad Soltany, Ruochen Deng, Shaowei Wang, Ximing Dong","submitted_at":"2026-06-12T20:46:04Z","abstract_excerpt":"LLMs have substantially improved software engineering yet real-world development requires architectural understanding. Such understanding is prohibitively expensive to label manually and impossible to verify through tests alone. We propose an agentic judging pipeline using a strong LLM as a scalable proxy for expert architectural evaluation, comprising two judges: the Architecture Complexity Judge (ACJ), which estimates codebase-specific architectural understanding a task demands, and the Architecture Quality Judge (AQJ), which evaluates patch conformance to repository-specific architectural c"},"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":"2606.14948","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-12T20:46:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"43fcdcfa99bb6f301657883655d6a8939d2c5a0f95ad3bf753172b323799bfce","abstract_canon_sha256":"463926777d5f71d90d3605a7261317787ef572b855740c0d7bd3bfe87fd8f1df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:31.496812Z","signature_b64":"7NdtTmL2cusyL3igZNNPbjq/BiZzsYklZyWG4tS8SeXkW+QAV2hYubVXith3BGCjIkfKhU8c5Kp2K3mesnq+AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5c98140148e3bc417852b564583196af49dda892392db0586eeb0be3bbac13c","last_reissued_at":"2026-07-08T01:18:31.496288Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:31.496288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Correctness: Enhancing Architectural Reasoning in Code LLMs via Scalable Labeling with Agentic Judgment","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Ahmed E. Hassan, Arthur Leung, Benjamin Rombaut, Boyuan Chen, Dayi Lin, Jiahuei Lin (Justina), Kirill Vasilevski, Milad Soltany, Ruochen Deng, Shaowei Wang, Ximing Dong","submitted_at":"2026-06-12T20:46:04Z","abstract_excerpt":"LLMs have substantially improved software engineering yet real-world development requires architectural understanding. Such understanding is prohibitively expensive to label manually and impossible to verify through tests alone. We propose an agentic judging pipeline using a strong LLM as a scalable proxy for expert architectural evaluation, comprising two judges: the Architecture Complexity Judge (ACJ), which estimates codebase-specific architectural understanding a task demands, and the Architecture Quality Judge (AQJ), which evaluates patch conformance to repository-specific architectural c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.14948","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/2606.14948/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":"2606.14948","created_at":"2026-07-08T01:18:31.496382+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.14948v2","created_at":"2026-07-08T01:18:31.496382+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.14948","created_at":"2026-07-08T01:18:31.496382+00:00"},{"alias_kind":"pith_short_12","alias_value":"6XEYCQAURY54","created_at":"2026-07-08T01:18:31.496382+00:00"},{"alias_kind":"pith_short_16","alias_value":"6XEYCQAURY54IF4F","created_at":"2026-07-08T01:18:31.496382+00:00"},{"alias_kind":"pith_short_8","alias_value":"6XEYCQAU","created_at":"2026-07-08T01:18:31.496382+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/6XEYCQAURY54IF4FFNLELAYZNL","json":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL.json","graph_json":"https://pith.science/api/pith-number/6XEYCQAURY54IF4FFNLELAYZNL/graph.json","events_json":"https://pith.science/api/pith-number/6XEYCQAURY54IF4FFNLELAYZNL/events.json","paper":"https://pith.science/paper/6XEYCQAU"},"agent_actions":{"view_html":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL","download_json":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL.json","view_paper":"https://pith.science/paper/6XEYCQAU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.14948&json=true","fetch_graph":"https://pith.science/api/pith-number/6XEYCQAURY54IF4FFNLELAYZNL/graph.json","fetch_events":"https://pith.science/api/pith-number/6XEYCQAURY54IF4FFNLELAYZNL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL/action/storage_attestation","attest_author":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL/action/author_attestation","sign_citation":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL/action/citation_signature","submit_replication":"https://pith.science/pith/6XEYCQAURY54IF4FFNLELAYZNL/action/replication_record"}},"created_at":"2026-07-08T01:18:31.496382+00:00","updated_at":"2026-07-08T01:18:31.496382+00:00"}