{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:JSXAQSNQBYPHOW3AMUQUYX5KPY","short_pith_number":"pith:JSXAQSNQ","schema_version":"1.0","canonical_sha256":"4cae0849b00e1e775b6065214c5faa7e2e9b23f558dd8ec2d5d24b77934665f7","source":{"kind":"arxiv","id":"2101.08543","version":2},"attestation_state":"computed","paper":{"title":"Boost then Convolve: Gradient Boosting Meets Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SI"],"primary_cat":"cs.LG","authors_text":"Liudmila Prokhorenkova, Sergei Ivanov","submitted_at":"2021-01-21T10:46:41Z","abstract_excerpt":"Graph neural networks (GNNs) are powerful models that have been successful in various graph representation learning tasks. Whereas gradient boosted decision trees (GBDT) often outperform other machine learning methods when faced with heterogeneous tabular data. But what approach should be used for graphs with tabular node features? Previous GNN models have mostly focused on networks with homogeneous sparse features and, as we show, are suboptimal in the heterogeneous setting. In this work, we propose a novel architecture that trains GBDT and GNN jointly to get the best of both worlds: the GBDT"},"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":"2101.08543","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-01-21T10:46:41Z","cross_cats_sorted":["cs.AI","cs.SI"],"title_canon_sha256":"d316e2fc3f534ac680816158e597eda5a616fcf38463bf595db2f927fff75458","abstract_canon_sha256":"4384e847521b0dcab2aa603b771ba698030188b6998162a1baa4d40ae70094fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:28:07.705239Z","signature_b64":"as/JQjvKmKCkF4heyD/HmJbqs1TjSHQI4eoSxQ7IbOowX7GuMiOc4fOa/fP1x4bqI16piF6rz6bONNftXsWYAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4cae0849b00e1e775b6065214c5faa7e2e9b23f558dd8ec2d5d24b77934665f7","last_reissued_at":"2026-07-05T02:28:07.704672Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:28:07.704672Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Boost then Convolve: Gradient Boosting Meets Graph Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SI"],"primary_cat":"cs.LG","authors_text":"Liudmila Prokhorenkova, Sergei Ivanov","submitted_at":"2021-01-21T10:46:41Z","abstract_excerpt":"Graph neural networks (GNNs) are powerful models that have been successful in various graph representation learning tasks. Whereas gradient boosted decision trees (GBDT) often outperform other machine learning methods when faced with heterogeneous tabular data. But what approach should be used for graphs with tabular node features? Previous GNN models have mostly focused on networks with homogeneous sparse features and, as we show, are suboptimal in the heterogeneous setting. In this work, we propose a novel architecture that trains GBDT and GNN jointly to get the best of both worlds: the GBDT"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.08543","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/2101.08543/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":"2101.08543","created_at":"2026-07-05T02:28:07.704754+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.08543v2","created_at":"2026-07-05T02:28:07.704754+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.08543","created_at":"2026-07-05T02:28:07.704754+00:00"},{"alias_kind":"pith_short_12","alias_value":"JSXAQSNQBYPH","created_at":"2026-07-05T02:28:07.704754+00:00"},{"alias_kind":"pith_short_16","alias_value":"JSXAQSNQBYPHOW3A","created_at":"2026-07-05T02:28:07.704754+00:00"},{"alias_kind":"pith_short_8","alias_value":"JSXAQSNQ","created_at":"2026-07-05T02:28:07.704754+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22429","citing_title":"Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation","ref_index":183,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12673","citing_title":"A Zero-shot Generalized Graph Anomaly Detection Framework via Node Reconstruction","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25429","citing_title":"Rethinking Feature Alignment in Generalist Graph Anomaly Detection: A Relational Fingerprint-based Approach","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2501.00309","citing_title":"Retrieval-Augmented Generation with Graphs (GraphRAG)","ref_index":171,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY","json":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY.json","graph_json":"https://pith.science/api/pith-number/JSXAQSNQBYPHOW3AMUQUYX5KPY/graph.json","events_json":"https://pith.science/api/pith-number/JSXAQSNQBYPHOW3AMUQUYX5KPY/events.json","paper":"https://pith.science/paper/JSXAQSNQ"},"agent_actions":{"view_html":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY","download_json":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY.json","view_paper":"https://pith.science/paper/JSXAQSNQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.08543&json=true","fetch_graph":"https://pith.science/api/pith-number/JSXAQSNQBYPHOW3AMUQUYX5KPY/graph.json","fetch_events":"https://pith.science/api/pith-number/JSXAQSNQBYPHOW3AMUQUYX5KPY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY/action/storage_attestation","attest_author":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY/action/author_attestation","sign_citation":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY/action/citation_signature","submit_replication":"https://pith.science/pith/JSXAQSNQBYPHOW3AMUQUYX5KPY/action/replication_record"}},"created_at":"2026-07-05T02:28:07.704754+00:00","updated_at":"2026-07-05T02:28:07.704754+00:00"}