{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QWR3KVZ6OCABVCWJUESKLLOQ7H","short_pith_number":"pith:QWR3KVZ6","schema_version":"1.0","canonical_sha256":"85a3b5573e70801a8ac9a124a5add0f9db4abbe0f0e3829a9bc19f8441c68f5b","source":{"kind":"arxiv","id":"2110.14081","version":3},"attestation_state":"computed","paper":{"title":"A Controlled Experiment of Different Code Representations for Learning-Based Bug Repair","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Ali Mesbah, Marjane Namavar, Noor Nashid","submitted_at":"2021-10-26T23:13:37Z","abstract_excerpt":"Training a deep learning model on source code has gained significant traction recently. Since such models reason about vectors of numbers, source code needs to be converted to a code representation before vectorization. Numerous approaches have been proposed to represent source code, from sequences of tokens to abstract syntax trees. However, there is no systematic study to understand the effect of code representation on learning performance. Through a controlled experiment, we examine the impact of various code representations on model accuracy and usefulness in deep learning-based program re"},"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":"2110.14081","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2021-10-26T23:13:37Z","cross_cats_sorted":[],"title_canon_sha256":"673a1f8ff1b5c20f6b7f447b43350f78ddade3db4d986187fe1da523b0bae98d","abstract_canon_sha256":"3ef8f6eca467f7c6854c1640fd716ddabff039c7dc0c21020f6eb31429cbd15d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:40:24.232333Z","signature_b64":"MQOPZKdlJBz00BRvpax948kRBsPZZzu/Riegzlpx9zR1rqD/uxYlPKMP3CHOKO+aazYIjaMlvVC++JOY8kOZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85a3b5573e70801a8ac9a124a5add0f9db4abbe0f0e3829a9bc19f8441c68f5b","last_reissued_at":"2026-07-05T04:40:24.231902Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:40:24.231902Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Controlled Experiment of Different Code Representations for Learning-Based Bug Repair","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Ali Mesbah, Marjane Namavar, Noor Nashid","submitted_at":"2021-10-26T23:13:37Z","abstract_excerpt":"Training a deep learning model on source code has gained significant traction recently. Since such models reason about vectors of numbers, source code needs to be converted to a code representation before vectorization. Numerous approaches have been proposed to represent source code, from sequences of tokens to abstract syntax trees. However, there is no systematic study to understand the effect of code representation on learning performance. Through a controlled experiment, we examine the impact of various code representations on model accuracy and usefulness in deep learning-based program re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.14081","kind":"arxiv","version":3},"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/2110.14081/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":"2110.14081","created_at":"2026-07-05T04:40:24.231957+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.14081v3","created_at":"2026-07-05T04:40:24.231957+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.14081","created_at":"2026-07-05T04:40:24.231957+00:00"},{"alias_kind":"pith_short_12","alias_value":"QWR3KVZ6OCAB","created_at":"2026-07-05T04:40:24.231957+00:00"},{"alias_kind":"pith_short_16","alias_value":"QWR3KVZ6OCABVCWJ","created_at":"2026-07-05T04:40:24.231957+00:00"},{"alias_kind":"pith_short_8","alias_value":"QWR3KVZ6","created_at":"2026-07-05T04:40:24.231957+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25356","citing_title":"Representation Matters: An Empirical Study of Program Representations for LLM Vulnerability Reasoning","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H","json":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H.json","graph_json":"https://pith.science/api/pith-number/QWR3KVZ6OCABVCWJUESKLLOQ7H/graph.json","events_json":"https://pith.science/api/pith-number/QWR3KVZ6OCABVCWJUESKLLOQ7H/events.json","paper":"https://pith.science/paper/QWR3KVZ6"},"agent_actions":{"view_html":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H","download_json":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H.json","view_paper":"https://pith.science/paper/QWR3KVZ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.14081&json=true","fetch_graph":"https://pith.science/api/pith-number/QWR3KVZ6OCABVCWJUESKLLOQ7H/graph.json","fetch_events":"https://pith.science/api/pith-number/QWR3KVZ6OCABVCWJUESKLLOQ7H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H/action/storage_attestation","attest_author":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H/action/author_attestation","sign_citation":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H/action/citation_signature","submit_replication":"https://pith.science/pith/QWR3KVZ6OCABVCWJUESKLLOQ7H/action/replication_record"}},"created_at":"2026-07-05T04:40:24.231957+00:00","updated_at":"2026-07-05T04:40:24.231957+00:00"}