{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3LCMPTRKHIGUKLWLAXNIK5PNSO","short_pith_number":"pith:3LCMPTRK","schema_version":"1.0","canonical_sha256":"dac4c7ce2a3a0d452ecb05da8575ed9385027c1612a83a8c7b8807c48f5552b0","source":{"kind":"arxiv","id":"2605.19369","version":1},"attestation_state":"computed","paper":{"title":"When to Answer and When to Defer: A Decision Framework for Reliable Code Predictions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Ravishka Rathnasuriya, Wei Yang","submitted_at":"2026-05-19T05:04:42Z","abstract_excerpt":"Code language models are increasingly adopted for both understanding and generative tasks. Despite their success, these models frequently produce overconfident incorrect predictions and underconfident correct predictions, undermining their reliability in deployment. Practical deployment demands three capabilities: accurately estimating the likelihood of correctness, abstaining on uncertain predictions, and invoking external mechanisms to validate or repair abstained outputs. Existing calibration and uncertainty estimation methods, primarily developed for natural language tasks, do not readily "},"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":"2605.19369","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-19T05:04:42Z","cross_cats_sorted":[],"title_canon_sha256":"b15e6fcfdb56bbb14163056efc90371c2b796e9742de5d616a5922d2a7e43ec4","abstract_canon_sha256":"6e8a40eebef0f7ae5f76b850fdc6abf89bb8003cd26e8a6969d4344540cd2b26"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:41.763044Z","signature_b64":"YaslGWkLL5Y6cg+Zd8n4RExWpiuHb24cMQAZWIsYP2JhwGahApy6W4a5gHXOWQow8u41CvVUVbaOnuYx+sDDBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dac4c7ce2a3a0d452ecb05da8575ed9385027c1612a83a8c7b8807c48f5552b0","last_reissued_at":"2026-05-20T01:05:41.762428Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:41.762428Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When to Answer and When to Defer: A Decision Framework for Reliable Code Predictions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Ravishka Rathnasuriya, Wei Yang","submitted_at":"2026-05-19T05:04:42Z","abstract_excerpt":"Code language models are increasingly adopted for both understanding and generative tasks. Despite their success, these models frequently produce overconfident incorrect predictions and underconfident correct predictions, undermining their reliability in deployment. Practical deployment demands three capabilities: accurately estimating the likelihood of correctness, abstaining on uncertain predictions, and invoking external mechanisms to validate or repair abstained outputs. Existing calibration and uncertainty estimation methods, primarily developed for natural language tasks, do not readily "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19369","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/2605.19369/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":"2605.19369","created_at":"2026-05-20T01:05:41.762555+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.19369v1","created_at":"2026-05-20T01:05:41.762555+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19369","created_at":"2026-05-20T01:05:41.762555+00:00"},{"alias_kind":"pith_short_12","alias_value":"3LCMPTRKHIGU","created_at":"2026-05-20T01:05:41.762555+00:00"},{"alias_kind":"pith_short_16","alias_value":"3LCMPTRKHIGUKLWL","created_at":"2026-05-20T01:05:41.762555+00:00"},{"alias_kind":"pith_short_8","alias_value":"3LCMPTRK","created_at":"2026-05-20T01:05:41.762555+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/3LCMPTRKHIGUKLWLAXNIK5PNSO","json":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO.json","graph_json":"https://pith.science/api/pith-number/3LCMPTRKHIGUKLWLAXNIK5PNSO/graph.json","events_json":"https://pith.science/api/pith-number/3LCMPTRKHIGUKLWLAXNIK5PNSO/events.json","paper":"https://pith.science/paper/3LCMPTRK"},"agent_actions":{"view_html":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO","download_json":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO.json","view_paper":"https://pith.science/paper/3LCMPTRK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.19369&json=true","fetch_graph":"https://pith.science/api/pith-number/3LCMPTRKHIGUKLWLAXNIK5PNSO/graph.json","fetch_events":"https://pith.science/api/pith-number/3LCMPTRKHIGUKLWLAXNIK5PNSO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO/action/storage_attestation","attest_author":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO/action/author_attestation","sign_citation":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO/action/citation_signature","submit_replication":"https://pith.science/pith/3LCMPTRKHIGUKLWLAXNIK5PNSO/action/replication_record"}},"created_at":"2026-05-20T01:05:41.762555+00:00","updated_at":"2026-05-20T01:05:41.762555+00:00"}