{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:37YFJDNFSDFU5RCHLFEPLKR2AZ","short_pith_number":"pith:37YFJDNF","schema_version":"1.0","canonical_sha256":"dff0548da590cb4ec4475948f5aa3a06520a6eb4c1f2a70f45a8816412a3f656","source":{"kind":"arxiv","id":"2212.02229","version":2},"attestation_state":"computed","paper":{"title":"GPS++: An Optimised Hybrid MPNN/Transformer for Molecular Property Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Adam Sanders, Deniz Beker, Dominic Masters, Dominique Beaini, Hatem Helal, Josef Dean, Kerstin Klaser, Ladislav Ramp\\'a\\v{s}ek, Sam Maddrell-Mander, Zhiyi Li","submitted_at":"2022-11-18T18:11:27Z","abstract_excerpt":"This technical report presents GPS++, the first-place solution to the Open Graph Benchmark Large-Scale Challenge (OGB-LSC 2022) for the PCQM4Mv2 molecular property prediction task. Our approach implements several key principles from the prior literature. At its core our GPS++ method is a hybrid MPNN/Transformer model that incorporates 3D atom positions and an auxiliary denoising task. The effectiveness of GPS++ is demonstrated by achieving 0.0719 mean absolute error on the independent test-challenge PCQM4Mv2 split. Thanks to Graphcore IPU acceleration, GPS++ scales to deep architectures (16 la"},"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":"2212.02229","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"q-bio.QM","submitted_at":"2022-11-18T18:11:27Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2e88a2f0e453ae0939981a7692a52aa3b14f6a17d8a3aaae8240e6e4139d0e0a","abstract_canon_sha256":"f5373aa4ddd0e4b5a6e908ed55de98d0577b0889d7662116979ac474a3b239a6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:23:02.396397Z","signature_b64":"+yym7Kjntst/J0KVt+e4kIDxAvU1b9BuC8irRK1B4cwRpd0xUBdlUCQtQ3/3Yy29HasN/IW4lYpx25FoTTQxDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dff0548da590cb4ec4475948f5aa3a06520a6eb4c1f2a70f45a8816412a3f656","last_reissued_at":"2026-07-05T05:23:02.395921Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:23:02.395921Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GPS++: An Optimised Hybrid MPNN/Transformer for Molecular Property Prediction","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Adam Sanders, Deniz Beker, Dominic Masters, Dominique Beaini, Hatem Helal, Josef Dean, Kerstin Klaser, Ladislav Ramp\\'a\\v{s}ek, Sam Maddrell-Mander, Zhiyi Li","submitted_at":"2022-11-18T18:11:27Z","abstract_excerpt":"This technical report presents GPS++, the first-place solution to the Open Graph Benchmark Large-Scale Challenge (OGB-LSC 2022) for the PCQM4Mv2 molecular property prediction task. Our approach implements several key principles from the prior literature. At its core our GPS++ method is a hybrid MPNN/Transformer model that incorporates 3D atom positions and an auxiliary denoising task. The effectiveness of GPS++ is demonstrated by achieving 0.0719 mean absolute error on the independent test-challenge PCQM4Mv2 split. Thanks to Graphcore IPU acceleration, GPS++ scales to deep architectures (16 la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2212.02229","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/2212.02229/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":"2212.02229","created_at":"2026-07-05T05:23:02.395977+00:00"},{"alias_kind":"arxiv_version","alias_value":"2212.02229v2","created_at":"2026-07-05T05:23:02.395977+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2212.02229","created_at":"2026-07-05T05:23:02.395977+00:00"},{"alias_kind":"pith_short_12","alias_value":"37YFJDNFSDFU","created_at":"2026-07-05T05:23:02.395977+00:00"},{"alias_kind":"pith_short_16","alias_value":"37YFJDNFSDFU5RCH","created_at":"2026-07-05T05:23:02.395977+00:00"},{"alias_kind":"pith_short_8","alias_value":"37YFJDNF","created_at":"2026-07-05T05:23:02.395977+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11508","citing_title":"Probabilistic Contrastive Pretraining for Multi-task ADME Property Prediction","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2602.17071","citing_title":"AdvSynGNN: Structure-Adaptive Graph Neural Nets via Adversarial Synthesis and Self-Corrective Propagation","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16586","citing_title":"A Systematic Survey and Benchmark of Deep Learning for Molecular Property Prediction in the Foundation Model Era","ref_index":101,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ","json":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ.json","graph_json":"https://pith.science/api/pith-number/37YFJDNFSDFU5RCHLFEPLKR2AZ/graph.json","events_json":"https://pith.science/api/pith-number/37YFJDNFSDFU5RCHLFEPLKR2AZ/events.json","paper":"https://pith.science/paper/37YFJDNF"},"agent_actions":{"view_html":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ","download_json":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ.json","view_paper":"https://pith.science/paper/37YFJDNF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2212.02229&json=true","fetch_graph":"https://pith.science/api/pith-number/37YFJDNFSDFU5RCHLFEPLKR2AZ/graph.json","fetch_events":"https://pith.science/api/pith-number/37YFJDNFSDFU5RCHLFEPLKR2AZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ/action/storage_attestation","attest_author":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ/action/author_attestation","sign_citation":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ/action/citation_signature","submit_replication":"https://pith.science/pith/37YFJDNFSDFU5RCHLFEPLKR2AZ/action/replication_record"}},"created_at":"2026-07-05T05:23:02.395977+00:00","updated_at":"2026-07-05T05:23:02.395977+00:00"}