{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:CBL7RZAJKMRWA3SMOC2UYTES5S","short_pith_number":"pith:CBL7RZAJ","schema_version":"1.0","canonical_sha256":"1057f8e4095323606e4c70b54c4c92eca68874e21678e03a0b45bfcca1ec7a2b","source":{"kind":"arxiv","id":"2106.08903","version":10},"attestation_state":"computed","paper":{"title":"GemNet: Universal Directional Graph Neural Networks for Molecules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","physics.chem-ph","stat.ML"],"primary_cat":"physics.comp-ph","authors_text":"Florian Becker, Johannes Gasteiger, Stephan G\\\"unnemann","submitted_at":"2021-06-02T15:44:55Z","abstract_excerpt":"Effectively predicting molecular interactions has the potential to accelerate molecular dynamics by multiple orders of magnitude and thus revolutionize chemical simulations. Graph neural networks (GNNs) have recently shown great successes for this task, overtaking classical methods based on fixed molecular kernels. However, they still appear very limited from a theoretical perspective, since regular GNNs cannot distinguish certain types of graphs. In this work we close this gap between theory and practice. We show that GNNs with spherical representations are indeed universal approximators for "},"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":"2106.08903","kind":"arxiv","version":10},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.comp-ph","submitted_at":"2021-06-02T15:44:55Z","cross_cats_sorted":["cs.LG","physics.chem-ph","stat.ML"],"title_canon_sha256":"1bb7e76bb9a5d1fbfec6b8b1f107ecbab77ac6d4d98a091e11a6248ff07800b0","abstract_canon_sha256":"2794b6b8df8772256b826a2d36c9aeeebea2feee32ed11d1333a7ae94c59cf34"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:35:32.708213Z","signature_b64":"WDQJr6rOk0E3c2hj/X1e86gG8btNf1sHt4hLjpo5vSV40W0BNkhzqeyPL8RCd/yBtIX753RvSZ1xyc473eQzCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1057f8e4095323606e4c70b54c4c92eca68874e21678e03a0b45bfcca1ec7a2b","last_reissued_at":"2026-07-05T08:35:32.707661Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:35:32.707661Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GemNet: Universal Directional Graph Neural Networks for Molecules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","physics.chem-ph","stat.ML"],"primary_cat":"physics.comp-ph","authors_text":"Florian Becker, Johannes Gasteiger, Stephan G\\\"unnemann","submitted_at":"2021-06-02T15:44:55Z","abstract_excerpt":"Effectively predicting molecular interactions has the potential to accelerate molecular dynamics by multiple orders of magnitude and thus revolutionize chemical simulations. Graph neural networks (GNNs) have recently shown great successes for this task, overtaking classical methods based on fixed molecular kernels. However, they still appear very limited from a theoretical perspective, since regular GNNs cannot distinguish certain types of graphs. In this work we close this gap between theory and practice. We show that GNNs with spherical representations are indeed universal approximators for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.08903","kind":"arxiv","version":10},"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/2106.08903/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":"2106.08903","created_at":"2026-07-05T08:35:32.707730+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.08903v10","created_at":"2026-07-05T08:35:32.707730+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.08903","created_at":"2026-07-05T08:35:32.707730+00:00"},{"alias_kind":"pith_short_12","alias_value":"CBL7RZAJKMRW","created_at":"2026-07-05T08:35:32.707730+00:00"},{"alias_kind":"pith_short_16","alias_value":"CBL7RZAJKMRWA3SM","created_at":"2026-07-05T08:35:32.707730+00:00"},{"alias_kind":"pith_short_8","alias_value":"CBL7RZAJ","created_at":"2026-07-05T08:35:32.707730+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29919","citing_title":"High-order tensor neural network for iteration-free structure relaxation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22698","citing_title":"Machine Learning Interatomic Potentials: Advancing Open-Source Software for Efficient and Scalable Molecular Simulation","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17147","citing_title":"Spatial statistics for screening molecular structures","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S","json":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S.json","graph_json":"https://pith.science/api/pith-number/CBL7RZAJKMRWA3SMOC2UYTES5S/graph.json","events_json":"https://pith.science/api/pith-number/CBL7RZAJKMRWA3SMOC2UYTES5S/events.json","paper":"https://pith.science/paper/CBL7RZAJ"},"agent_actions":{"view_html":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S","download_json":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S.json","view_paper":"https://pith.science/paper/CBL7RZAJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.08903&json=true","fetch_graph":"https://pith.science/api/pith-number/CBL7RZAJKMRWA3SMOC2UYTES5S/graph.json","fetch_events":"https://pith.science/api/pith-number/CBL7RZAJKMRWA3SMOC2UYTES5S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S/action/storage_attestation","attest_author":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S/action/author_attestation","sign_citation":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S/action/citation_signature","submit_replication":"https://pith.science/pith/CBL7RZAJKMRWA3SMOC2UYTES5S/action/replication_record"}},"created_at":"2026-07-05T08:35:32.707730+00:00","updated_at":"2026-07-05T08:35:32.707730+00:00"}