{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RJLGOLEUMJB2EUASALZ7SWWXSG","short_pith_number":"pith:RJLGOLEU","schema_version":"1.0","canonical_sha256":"8a56672c946243a2501202f3f95ad7918813cb72367ec02611bf9eaa0fac13ac","source":{"kind":"arxiv","id":"2302.11640","version":2},"attestation_state":"computed","paper":{"title":"A critical look at the evaluation of GNNs under heterophily: Are we really making progress?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Artem Babenko, Denis Kuznedelev, Liudmila Prokhorenkova, Michael Diskin, Oleg Platonov","submitted_at":"2023-02-22T20:32:59Z","abstract_excerpt":"Node classification is a classical graph machine learning task on which Graph Neural Networks (GNNs) have recently achieved strong results. However, it is often believed that standard GNNs only work well for homophilous graphs, i.e., graphs where edges tend to connect nodes of the same class. Graphs without this property are called heterophilous, and it is typically assumed that specialized methods are required to achieve strong performance on such graphs. In this work, we challenge this assumption. First, we show that the standard datasets used for evaluating heterophily-specific models have "},"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":"2302.11640","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-02-22T20:32:59Z","cross_cats_sorted":[],"title_canon_sha256":"e46d1189bee70ca82ebe7f9ddcbbca06b63da552ead4dac135b88c95a0a90641","abstract_canon_sha256":"23732dac5b1c97d40e363bfedd395b0022afa2a421bf3989060a6af7f5d48ffd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:26.455349Z","signature_b64":"NmdGis+crfeQbnhJGcADTRwOQk5Dc92jhD5lLsaNow47ugUnNTyt/tMa53Eb8YNoN+JpapMJw7yeU9EMLS2vBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a56672c946243a2501202f3f95ad7918813cb72367ec02611bf9eaa0fac13ac","last_reissued_at":"2026-07-05T07:51:26.454861Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:26.454861Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A critical look at the evaluation of GNNs under heterophily: Are we really making progress?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Artem Babenko, Denis Kuznedelev, Liudmila Prokhorenkova, Michael Diskin, Oleg Platonov","submitted_at":"2023-02-22T20:32:59Z","abstract_excerpt":"Node classification is a classical graph machine learning task on which Graph Neural Networks (GNNs) have recently achieved strong results. However, it is often believed that standard GNNs only work well for homophilous graphs, i.e., graphs where edges tend to connect nodes of the same class. Graphs without this property are called heterophilous, and it is typically assumed that specialized methods are required to achieve strong performance on such graphs. In this work, we challenge this assumption. First, we show that the standard datasets used for evaluating heterophily-specific models have "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.11640","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/2302.11640/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":"2302.11640","created_at":"2026-07-05T07:51:26.454920+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.11640v2","created_at":"2026-07-05T07:51:26.454920+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.11640","created_at":"2026-07-05T07:51:26.454920+00:00"},{"alias_kind":"pith_short_12","alias_value":"RJLGOLEUMJB2","created_at":"2026-07-05T07:51:26.454920+00:00"},{"alias_kind":"pith_short_16","alias_value":"RJLGOLEUMJB2EUAS","created_at":"2026-07-05T07:51:26.454920+00:00"},{"alias_kind":"pith_short_8","alias_value":"RJLGOLEU","created_at":"2026-07-05T07:51:26.454920+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24958","citing_title":"Swarm-Inspired Generation of Collective Behaviors in Graph Dynamical Systems","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01660","citing_title":"Gate the Filter, Not the Message: Node-Channel Mixtures for Pre-Propagation GNNs","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12827","citing_title":"GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21247","citing_title":"Graph Navier Stokes Networks","ref_index":65,"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":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26857","citing_title":"Generalist Graph Anomaly Detection via Prototype-Based Distillation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31371","citing_title":"Softsign: Smooth Sign in Your Optimizer For Better Parameter Heterogeneity Handling","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20248","citing_title":"Graph Transductive Sharpening: Leveraging Unlabeled Predictions in Node Classification","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20879","citing_title":"NeighborDiv: Training-free Zero-shot Generalist Graph Anomaly Detection via Neighbor Diversity","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21435","citing_title":"Gaussian Sheaf Neural Networks","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12827","citing_title":"GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10975","citing_title":"Hierarchical Multi-Scale Graph Neural Networks: Scalable Heterophilous Learning with Oversmoothing and Oversquashing Mitigation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11987","citing_title":"Random-Set Graph Neural Networks","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27387","citing_title":"Robust Learning on Heterogeneous Graphs with Heterophily: A Graph Structure Learning Approach","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09085","citing_title":"Beyond Isolated Clients: Integrating Graph-Based Embeddings into Event Sequence Models","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG","json":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG.json","graph_json":"https://pith.science/api/pith-number/RJLGOLEUMJB2EUASALZ7SWWXSG/graph.json","events_json":"https://pith.science/api/pith-number/RJLGOLEUMJB2EUASALZ7SWWXSG/events.json","paper":"https://pith.science/paper/RJLGOLEU"},"agent_actions":{"view_html":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG","download_json":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG.json","view_paper":"https://pith.science/paper/RJLGOLEU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.11640&json=true","fetch_graph":"https://pith.science/api/pith-number/RJLGOLEUMJB2EUASALZ7SWWXSG/graph.json","fetch_events":"https://pith.science/api/pith-number/RJLGOLEUMJB2EUASALZ7SWWXSG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG/action/storage_attestation","attest_author":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG/action/author_attestation","sign_citation":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG/action/citation_signature","submit_replication":"https://pith.science/pith/RJLGOLEUMJB2EUASALZ7SWWXSG/action/replication_record"}},"created_at":"2026-07-05T07:51:26.454920+00:00","updated_at":"2026-07-05T07:51:26.454920+00:00"}