{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4LFMY7YZHCWDTVMT652JHNXGU6","short_pith_number":"pith:4LFMY7YZ","schema_version":"1.0","canonical_sha256":"e2cacc7f1938ac39d593f77493b6e6a7b5e07ce96aad01cabcdafd0209904dbd","source":{"kind":"arxiv","id":"2605.22133","version":1},"attestation_state":"computed","paper":{"title":"Atom-level Protein Representation Learning Improves Protein Structure Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"q-bio.BM","authors_text":"Hyeongwoo Kim, Hyosoon Jang, Hyunjin Seo, Mingyeong Shin, Seonghwan Seo, Sungsoo Ahn, Taewon Kim, Wonho Zhung, Wooyoun Kim","submitted_at":"2026-05-21T08:07:36Z","abstract_excerpt":"Recent advances in generative modeling show that pretrained representations can improve generation as conditioning features or alignment targets. Motivated by this, we study protein representations for predicting structures beyond conventional function annotation. We propose TriProRep, a structure-aware pretraining method that jointly models three aligned residue-level views: amino-acid identity, backbone geometry, and local full-atom geometry, discretely encoded via VQ-VAE tokenizers. By pretraining to recover original tokens from generator-corrupted views, TriProRep learns to distinguish pla"},"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":"2605.22133","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"q-bio.BM","submitted_at":"2026-05-21T08:07:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"aef4b3fcf5a4ddf547fa50d74777e3349c7f72dc7cb1c024b82e6b3f719080e7","abstract_canon_sha256":"b6432bf34d806735d6d5997a2e7fc8a9604841b1d5cfa8c372d352051f61af69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:27.601731Z","signature_b64":"WznpdAg+SaGLqS6B9nfIuRVikNsYG8W4SVoW4QCcrX7fjZS3PevKDXMwDYV/wiyJYiQA2NRIsVXxy++tzMzfAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e2cacc7f1938ac39d593f77493b6e6a7b5e07ce96aad01cabcdafd0209904dbd","last_reissued_at":"2026-05-22T01:04:27.600781Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:27.600781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Atom-level Protein Representation Learning Improves Protein Structure Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"q-bio.BM","authors_text":"Hyeongwoo Kim, Hyosoon Jang, Hyunjin Seo, Mingyeong Shin, Seonghwan Seo, Sungsoo Ahn, Taewon Kim, Wonho Zhung, Wooyoun Kim","submitted_at":"2026-05-21T08:07:36Z","abstract_excerpt":"Recent advances in generative modeling show that pretrained representations can improve generation as conditioning features or alignment targets. Motivated by this, we study protein representations for predicting structures beyond conventional function annotation. We propose TriProRep, a structure-aware pretraining method that jointly models three aligned residue-level views: amino-acid identity, backbone geometry, and local full-atom geometry, discretely encoded via VQ-VAE tokenizers. By pretraining to recover original tokens from generator-corrupted views, TriProRep learns to distinguish pla"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22133","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/2605.22133/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":"2605.22133","created_at":"2026-05-22T01:04:27.600952+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.22133v1","created_at":"2026-05-22T01:04:27.600952+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22133","created_at":"2026-05-22T01:04:27.600952+00:00"},{"alias_kind":"pith_short_12","alias_value":"4LFMY7YZHCWD","created_at":"2026-05-22T01:04:27.600952+00:00"},{"alias_kind":"pith_short_16","alias_value":"4LFMY7YZHCWDTVMT","created_at":"2026-05-22T01:04:27.600952+00:00"},{"alias_kind":"pith_short_8","alias_value":"4LFMY7YZ","created_at":"2026-05-22T01:04:27.600952+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/4LFMY7YZHCWDTVMT652JHNXGU6","json":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6.json","graph_json":"https://pith.science/api/pith-number/4LFMY7YZHCWDTVMT652JHNXGU6/graph.json","events_json":"https://pith.science/api/pith-number/4LFMY7YZHCWDTVMT652JHNXGU6/events.json","paper":"https://pith.science/paper/4LFMY7YZ"},"agent_actions":{"view_html":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6","download_json":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6.json","view_paper":"https://pith.science/paper/4LFMY7YZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.22133&json=true","fetch_graph":"https://pith.science/api/pith-number/4LFMY7YZHCWDTVMT652JHNXGU6/graph.json","fetch_events":"https://pith.science/api/pith-number/4LFMY7YZHCWDTVMT652JHNXGU6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6/action/storage_attestation","attest_author":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6/action/author_attestation","sign_citation":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6/action/citation_signature","submit_replication":"https://pith.science/pith/4LFMY7YZHCWDTVMT652JHNXGU6/action/replication_record"}},"created_at":"2026-05-22T01:04:27.600952+00:00","updated_at":"2026-05-22T01:04:27.600952+00:00"}