{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RJMX6LWZPATVXAWN7NPODGY47F","short_pith_number":"pith:RJMX6LWZ","schema_version":"1.0","canonical_sha256":"8a597f2ed978275b82cdfb5ee19b1cf95029314dd9b23d469ee306216fc70497","source":{"kind":"arxiv","id":"2606.19741","version":1},"attestation_state":"computed","paper":{"title":"Interpreting Neural Combinatorial Optimization via Evolving Programmatic Bottlenecks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anqi Xie, Cathy Wu, Haocheng Duan, Jieyi Bi, Sirui Li, Yining Ma, Yuxin Guo","submitted_at":"2026-06-18T03:14:10Z","abstract_excerpt":"Neural Combinatorial Optimization (NCO) achieves strong performance, yet its black-box nature remains a key roadblock to deployment and scientific diagnosis. Standard interpretability tools, such as Concept Bottleneck Models (CBMs), are ill-equipped for NCO, whose decisions are dynamic, state-dependent, and lack proper concept vocabulary definition. To close this gap, we introduce Evolving Programmatic Bottlenecks (EPB), to our knowledge, the first framework for interpreting NCO policies by distilling black-box NCO models into human-readable program portfolios. EPB employs an LLM to autonomous"},"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":"2606.19741","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-18T03:14:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7d735a18e35971579e631efa9b70916df79fbc1cf4911edb567673936f62fc6b","abstract_canon_sha256":"4568754de05cd2a570d5d4ef0d694651ef4a0aefe977b49cc1a34dbf86f17451"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:33.825233Z","signature_b64":"b7VbOnqNL95oWq3/gAWb2tvAPkTZFESkSlteYeJhYj67vGPXzqDJjRURXQ9liLNq8qSoFG6uRbY2XRYfC567AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a597f2ed978275b82cdfb5ee19b1cf95029314dd9b23d469ee306216fc70497","last_reissued_at":"2026-06-19T16:12:33.824877Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:33.824877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Interpreting Neural Combinatorial Optimization via Evolving Programmatic Bottlenecks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Anqi Xie, Cathy Wu, Haocheng Duan, Jieyi Bi, Sirui Li, Yining Ma, Yuxin Guo","submitted_at":"2026-06-18T03:14:10Z","abstract_excerpt":"Neural Combinatorial Optimization (NCO) achieves strong performance, yet its black-box nature remains a key roadblock to deployment and scientific diagnosis. Standard interpretability tools, such as Concept Bottleneck Models (CBMs), are ill-equipped for NCO, whose decisions are dynamic, state-dependent, and lack proper concept vocabulary definition. To close this gap, we introduce Evolving Programmatic Bottlenecks (EPB), to our knowledge, the first framework for interpreting NCO policies by distilling black-box NCO models into human-readable program portfolios. EPB employs an LLM to autonomous"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19741","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/2606.19741/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":"2606.19741","created_at":"2026-06-19T16:12:33.824933+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19741v1","created_at":"2026-06-19T16:12:33.824933+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19741","created_at":"2026-06-19T16:12:33.824933+00:00"},{"alias_kind":"pith_short_12","alias_value":"RJMX6LWZPATV","created_at":"2026-06-19T16:12:33.824933+00:00"},{"alias_kind":"pith_short_16","alias_value":"RJMX6LWZPATVXAWN","created_at":"2026-06-19T16:12:33.824933+00:00"},{"alias_kind":"pith_short_8","alias_value":"RJMX6LWZ","created_at":"2026-06-19T16:12:33.824933+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/RJMX6LWZPATVXAWN7NPODGY47F","json":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F.json","graph_json":"https://pith.science/api/pith-number/RJMX6LWZPATVXAWN7NPODGY47F/graph.json","events_json":"https://pith.science/api/pith-number/RJMX6LWZPATVXAWN7NPODGY47F/events.json","paper":"https://pith.science/paper/RJMX6LWZ"},"agent_actions":{"view_html":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F","download_json":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F.json","view_paper":"https://pith.science/paper/RJMX6LWZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19741&json=true","fetch_graph":"https://pith.science/api/pith-number/RJMX6LWZPATVXAWN7NPODGY47F/graph.json","fetch_events":"https://pith.science/api/pith-number/RJMX6LWZPATVXAWN7NPODGY47F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F/action/storage_attestation","attest_author":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F/action/author_attestation","sign_citation":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F/action/citation_signature","submit_replication":"https://pith.science/pith/RJMX6LWZPATVXAWN7NPODGY47F/action/replication_record"}},"created_at":"2026-06-19T16:12:33.824933+00:00","updated_at":"2026-06-19T16:12:33.824933+00:00"}