{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:R6PUELSFBAPWG2NQJJAGINPYGK","short_pith_number":"pith:R6PUELSF","schema_version":"1.0","canonical_sha256":"8f9f422e45081f6369b04a406435f83287e89d70895cb2fa55f04a8f069a05ab","source":{"kind":"arxiv","id":"2411.17538","version":2},"attestation_state":"computed","paper":{"title":"Isotropy Matters: Soft-ZCA Whitening of Embeddings for Semantic Code Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andor Diera, Ansgar Scherp, Lukas Galke","submitted_at":"2024-11-26T15:53:28Z","abstract_excerpt":"Low isotropy in an embedding space impairs performance on tasks involving semantic inference. Our study investigates the impact of isotropy on semantic code search performance and explores post-processing techniques to mitigate this issue. We analyze various code language models, examine isotropy in their embedding spaces, and its influence on search effectiveness. We propose a modified ZCA whitening technique to control isotropy levels in embeddings. Our results demonstrate that Soft-ZCA whitening improves the performance of pre-trained code language models and can complement contrastive fine"},"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":"2411.17538","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-26T15:53:28Z","cross_cats_sorted":[],"title_canon_sha256":"d7859c9fd46c78ca1841382b563f71884dcfb0429b3edb151872d88f5ed40f5e","abstract_canon_sha256":"623b8411e0b0942345f1d9d1ee856ebb2e0498cfbb6e24b902e92b4c1f98aebb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:05.029078Z","signature_b64":"hOAMpq7yH2Na+gyM5eovMkRsjOV/ZAEQVv324kJk+DtSEAN1mmTZEqN0TvKaAAzvttFSNwuK2ESlIkS1RQTJAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f9f422e45081f6369b04a406435f83287e89d70895cb2fa55f04a8f069a05ab","last_reissued_at":"2026-07-05T09:41:05.028585Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:05.028585Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Isotropy Matters: Soft-ZCA Whitening of Embeddings for Semantic Code Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andor Diera, Ansgar Scherp, Lukas Galke","submitted_at":"2024-11-26T15:53:28Z","abstract_excerpt":"Low isotropy in an embedding space impairs performance on tasks involving semantic inference. Our study investigates the impact of isotropy on semantic code search performance and explores post-processing techniques to mitigate this issue. We analyze various code language models, examine isotropy in their embedding spaces, and its influence on search effectiveness. We propose a modified ZCA whitening technique to control isotropy levels in embeddings. Our results demonstrate that Soft-ZCA whitening improves the performance of pre-trained code language models and can complement contrastive fine"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17538","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/2411.17538/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":"2411.17538","created_at":"2026-07-05T09:41:05.028643+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.17538v2","created_at":"2026-07-05T09:41:05.028643+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17538","created_at":"2026-07-05T09:41:05.028643+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6PUELSFBAPW","created_at":"2026-07-05T09:41:05.028643+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6PUELSFBAPWG2NQ","created_at":"2026-07-05T09:41:05.028643+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6PUELSF","created_at":"2026-07-05T09:41:05.028643+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31602","citing_title":"Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31602","citing_title":"Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK","json":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK.json","graph_json":"https://pith.science/api/pith-number/R6PUELSFBAPWG2NQJJAGINPYGK/graph.json","events_json":"https://pith.science/api/pith-number/R6PUELSFBAPWG2NQJJAGINPYGK/events.json","paper":"https://pith.science/paper/R6PUELSF"},"agent_actions":{"view_html":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK","download_json":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK.json","view_paper":"https://pith.science/paper/R6PUELSF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.17538&json=true","fetch_graph":"https://pith.science/api/pith-number/R6PUELSFBAPWG2NQJJAGINPYGK/graph.json","fetch_events":"https://pith.science/api/pith-number/R6PUELSFBAPWG2NQJJAGINPYGK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK/action/storage_attestation","attest_author":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK/action/author_attestation","sign_citation":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK/action/citation_signature","submit_replication":"https://pith.science/pith/R6PUELSFBAPWG2NQJJAGINPYGK/action/replication_record"}},"created_at":"2026-07-05T09:41:05.028643+00:00","updated_at":"2026-07-05T09:41:05.028643+00:00"}