{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4ODVX2VTX6HWCLUQ5RL2WNQIKE","short_pith_number":"pith:4ODVX2VT","schema_version":"1.0","canonical_sha256":"e3875beab3bf8f612e90ec57ab36085133cba702643b9668194ec27b0220efaa","source":{"kind":"arxiv","id":"2308.03312","version":9},"attestation_state":"computed","paper":{"title":"Exploiting Code Symmetries for Learning Program Semantics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.PL"],"primary_cat":"cs.LG","authors_text":"Junfeng Yang, Kexin Pei, Lorenzo Cavallaro, Qirui Jin, Scott Geng, Shuyang Liu, Suman Jana, Weichen Li","submitted_at":"2023-08-07T05:40:58Z","abstract_excerpt":"This paper tackles the challenge of teaching code semantics to Large Language Models (LLMs) for program analysis by incorporating code symmetries into the model architecture. We introduce a group-theoretic framework that defines code symmetries as semantics-preserving transformations, where forming a code symmetry group enables precise and efficient reasoning of code semantics. Our solution, SymC, develops a novel variant of self-attention that is provably equivariant to code symmetries from the permutation group defined over the program dependence graph. SymC obtains superior performance on f"},"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":"2308.03312","kind":"arxiv","version":9},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-08-07T05:40:58Z","cross_cats_sorted":["cs.CR","cs.PL"],"title_canon_sha256":"65c1811ee3ef165b79fe58697db9dbed64e23cd7d69371f3d01a2f8cd3b164bc","abstract_canon_sha256":"f0949590392d145298262726cbfef5025213ce6a510e9c247c6b90f43d68e058"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:04:28.888848Z","signature_b64":"RC++JgSKj5/iD5QpMUfdqXhC1bieeLzfEYHry70h38MNDTkgBlw07zCrz3gjYllauktiHxecvLPQjwb1EZoMDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3875beab3bf8f612e90ec57ab36085133cba702643b9668194ec27b0220efaa","last_reissued_at":"2026-07-05T09:04:28.888372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:04:28.888372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploiting Code Symmetries for Learning Program Semantics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.PL"],"primary_cat":"cs.LG","authors_text":"Junfeng Yang, Kexin Pei, Lorenzo Cavallaro, Qirui Jin, Scott Geng, Shuyang Liu, Suman Jana, Weichen Li","submitted_at":"2023-08-07T05:40:58Z","abstract_excerpt":"This paper tackles the challenge of teaching code semantics to Large Language Models (LLMs) for program analysis by incorporating code symmetries into the model architecture. We introduce a group-theoretic framework that defines code symmetries as semantics-preserving transformations, where forming a code symmetry group enables precise and efficient reasoning of code semantics. Our solution, SymC, develops a novel variant of self-attention that is provably equivariant to code symmetries from the permutation group defined over the program dependence graph. SymC obtains superior performance on f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.03312","kind":"arxiv","version":9},"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/2308.03312/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":"2308.03312","created_at":"2026-07-05T09:04:28.888431+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.03312v9","created_at":"2026-07-05T09:04:28.888431+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.03312","created_at":"2026-07-05T09:04:28.888431+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ODVX2VTX6HW","created_at":"2026-07-05T09:04:28.888431+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ODVX2VTX6HWCLUQ","created_at":"2026-07-05T09:04:28.888431+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ODVX2VT","created_at":"2026-07-05T09:04:28.888431+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.14399","citing_title":"NESA: Relational Neuro-Symbolic Static Program Analysis","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE","json":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE.json","graph_json":"https://pith.science/api/pith-number/4ODVX2VTX6HWCLUQ5RL2WNQIKE/graph.json","events_json":"https://pith.science/api/pith-number/4ODVX2VTX6HWCLUQ5RL2WNQIKE/events.json","paper":"https://pith.science/paper/4ODVX2VT"},"agent_actions":{"view_html":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE","download_json":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE.json","view_paper":"https://pith.science/paper/4ODVX2VT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.03312&json=true","fetch_graph":"https://pith.science/api/pith-number/4ODVX2VTX6HWCLUQ5RL2WNQIKE/graph.json","fetch_events":"https://pith.science/api/pith-number/4ODVX2VTX6HWCLUQ5RL2WNQIKE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE/action/storage_attestation","attest_author":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE/action/author_attestation","sign_citation":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE/action/citation_signature","submit_replication":"https://pith.science/pith/4ODVX2VTX6HWCLUQ5RL2WNQIKE/action/replication_record"}},"created_at":"2026-07-05T09:04:28.888431+00:00","updated_at":"2026-07-05T09:04:28.888431+00:00"}