{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:B4GAKXG7DDJLHSLGMF6VF7ZWLT","short_pith_number":"pith:B4GAKXG7","schema_version":"1.0","canonical_sha256":"0f0c055cdf18d2b3c966617d52ff365cfd0f64fe8309d83cb520c026aa97c83f","source":{"kind":"arxiv","id":"2605.16785","version":1},"attestation_state":"computed","paper":{"title":"Encoding Robust Topological Signatures for Hyperdimensional Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Topology-guided hyperdimensional computing resists image corruptions by encoding holes and rotation-invariant shape signatures.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Arpan Kusari","submitted_at":"2026-05-16T03:36:40Z","abstract_excerpt":"Hyperdimensional (HD) computing offers an attractive alternative to deep networks for edge learning due to its simplicity, fast prototype-based inference, and compatibility with online updates. However, standard pixel-based HD encoders are brittle: small distribution shifts such as rotation, noise, or occlusion can drastically reduce accuracy. We extract discrete topological primitives-most notably holes-from binarized shapes and pair them with rotation/translation/scale (RTS)-invariant shape signatures. Our method constructs RTS-stable descriptors for (i) the outer shape using a spatial-pyram"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.16785","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T03:36:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d33c6fa9025dccc531202d4bce85d9411fb1cb65975743e14f3c191da3cdd314","abstract_canon_sha256":"d4cd5c7cf2470a773e15050012182d663b3a16853b08463fa2822178d2756894"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:21.873279Z","signature_b64":"aMU93p1zqgBWTHuWazydjgCWDBMBw6mkpMPk5Gi38UHSm4GtuGF01os1ojQXGxa3rNqd56inl1jKv5hdybe9Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f0c055cdf18d2b3c966617d52ff365cfd0f64fe8309d83cb520c026aa97c83f","last_reissued_at":"2026-05-20T00:03:21.872417Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:21.872417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Encoding Robust Topological Signatures for Hyperdimensional Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Topology-guided hyperdimensional computing resists image corruptions by encoding holes and rotation-invariant shape signatures.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Arpan Kusari","submitted_at":"2026-05-16T03:36:40Z","abstract_excerpt":"Hyperdimensional (HD) computing offers an attractive alternative to deep networks for edge learning due to its simplicity, fast prototype-based inference, and compatibility with online updates. However, standard pixel-based HD encoders are brittle: small distribution shifts such as rotation, noise, or occlusion can drastically reduce accuracy. We extract discrete topological primitives-most notably holes-from binarized shapes and pair them with rotation/translation/scale (RTS)-invariant shape signatures. Our method constructs RTS-stable descriptors for (i) the outer shape using a spatial-pyram"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Topology-guided HD computing substantially improves robustness compared with a naive HD baseline, maintaining high accuracy across multiple corruption families and benefiting from lightweight online training. Compared with a compact CNN trained on clean data, our method achieves competitive clean accuracy while offering markedly stronger robustness to several pixel-level corruptions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That holes and other topological primitives can be reliably extracted from binarized shapes even under the tested corruptions (rotation, Gaussian noise, salt-and-pepper, cutout, zoom) and that the chosen RTS-invariant descriptors (spatial-pyramid Zernike and intrinsic Fourier radial signatures) preserve the information needed for discrimination.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Topology-guided HD computing encodes discrete holes and RTS-invariant descriptors (Zernike for outer shape, Fourier for holes) into hypervectors with learned reliability weights, yielding substantially higher robustness on corrupted MNIST/EMNIST than naive HD baselines while matching compact CNNs on","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Topology-guided hyperdimensional computing resists image corruptions by encoding holes and rotation-invariant shape signatures.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d9a657694e938dcc6110d3b57c57f1c6729e5f3ccfa2ecb9ba052a428702f6df"},"source":{"id":"2605.16785","kind":"arxiv","version":1},"verdict":{"id":"85192589-ebc6-4f2f-b467-68e28e158959","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T21:33:20.804640Z","strongest_claim":"Topology-guided HD computing substantially improves robustness compared with a naive HD baseline, maintaining high accuracy across multiple corruption families and benefiting from lightweight online training. Compared with a compact CNN trained on clean data, our method achieves competitive clean accuracy while offering markedly stronger robustness to several pixel-level corruptions.","one_line_summary":"Topology-guided HD computing encodes discrete holes and RTS-invariant descriptors (Zernike for outer shape, Fourier for holes) into hypervectors with learned reliability weights, yielding substantially higher robustness on corrupted MNIST/EMNIST than naive HD baselines while matching compact CNNs on","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That holes and other topological primitives can be reliably extracted from binarized shapes even under the tested corruptions (rotation, Gaussian noise, salt-and-pepper, cutout, zoom) and that the chosen RTS-invariant descriptors (spatial-pyramid Zernike and intrinsic Fourier radial signatures) preserve the information needed for discrimination.","pith_extraction_headline":"Topology-guided hyperdimensional computing resists image corruptions by encoding holes and rotation-invariant shape signatures."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16785/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:19.660713Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T21:40:53.373255Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.299231Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.434916Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"666a2c81beeb48ae792f2dd5aae28210c809c93c45e7585c707c4edfd0f8abe7"},"references":{"count":18,"sample":[{"doi":"","year":2024,"title":"Frontiers in big data , volume=","work_id":"e878a96f-a80f-46f1-8250-88177526b32f","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"2025 11th International Conference on Computing and Artificial Intelligence (ICCAI) , pages=","work_id":"20681bc7-d65a-4549-81ac-4536d6efc138","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Journal of Artificial Intelligence Research , volume=","work_id":"6f9167c2-adcb-4b88-9814-3067d4e82af5","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2009,"title":"Cognitive computation , volume=","work_id":"e1385c2f-3b87-4682-8a6c-927830a703f3","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"ACM Computing Surveys , volume=","work_id":"484d6600-fc97-4ca4-a8a4-467a67075e51","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":18,"snapshot_sha256":"f45b6b2a566fa3ed0cd323bad9ade358096c71eafcd64d20f12da3b973a829be","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"3bc8c31239dd9f131c7ed794e77f4d70be5ae9abf6c6976879d96eee59baac52"},"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":"2605.16785","created_at":"2026-05-20T00:03:21.872553+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16785v1","created_at":"2026-05-20T00:03:21.872553+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16785","created_at":"2026-05-20T00:03:21.872553+00:00"},{"alias_kind":"pith_short_12","alias_value":"B4GAKXG7DDJL","created_at":"2026-05-20T00:03:21.872553+00:00"},{"alias_kind":"pith_short_16","alias_value":"B4GAKXG7DDJLHSLG","created_at":"2026-05-20T00:03:21.872553+00:00"},{"alias_kind":"pith_short_8","alias_value":"B4GAKXG7","created_at":"2026-05-20T00:03:21.872553+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":1,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT","json":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT.json","graph_json":"https://pith.science/api/pith-number/B4GAKXG7DDJLHSLGMF6VF7ZWLT/graph.json","events_json":"https://pith.science/api/pith-number/B4GAKXG7DDJLHSLGMF6VF7ZWLT/events.json","paper":"https://pith.science/paper/B4GAKXG7"},"agent_actions":{"view_html":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT","download_json":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT.json","view_paper":"https://pith.science/paper/B4GAKXG7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16785&json=true","fetch_graph":"https://pith.science/api/pith-number/B4GAKXG7DDJLHSLGMF6VF7ZWLT/graph.json","fetch_events":"https://pith.science/api/pith-number/B4GAKXG7DDJLHSLGMF6VF7ZWLT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT/action/storage_attestation","attest_author":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT/action/author_attestation","sign_citation":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT/action/citation_signature","submit_replication":"https://pith.science/pith/B4GAKXG7DDJLHSLGMF6VF7ZWLT/action/replication_record"}},"created_at":"2026-05-20T00:03:21.872553+00:00","updated_at":"2026-05-20T00:03:21.872553+00:00"}