{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:RKFAU3I22SLFGCP5NEMI3E42GM","short_pith_number":"pith:RKFAU3I2","schema_version":"1.0","canonical_sha256":"8a8a0a6d1ad4965309fd69188d939a331b984a83c6ac8fb6b37cade408c962a8","source":{"kind":"arxiv","id":"2010.15116","version":2},"attestation_state":"computed","paper":{"title":"On Graph Neural Networks versus Graph-Augmented MLPs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.CO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Joan Bruna, Lei Chen, Zhengdao Chen","submitted_at":"2020-10-28T17:59:59Z","abstract_excerpt":"From the perspective of expressive power, this work compares multi-layer Graph Neural Networks (GNNs) with a simplified alternative that we call Graph-Augmented Multi-Layer Perceptrons (GA-MLPs), which first augments node features with certain multi-hop operators on the graph and then applies an MLP in a node-wise fashion. From the perspective of graph isomorphism testing, we show both theoretically and numerically that GA-MLPs with suitable operators can distinguish almost all non-isomorphic graphs, just like the Weifeiler-Lehman (WL) test. However, by viewing them as node-level functions and"},"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":"2010.15116","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-28T17:59:59Z","cross_cats_sorted":["math.CO","stat.ML"],"title_canon_sha256":"5aa980b7fa705a60dc5c1f28d36488c702265bf62a9951301496045edc79d60a","abstract_canon_sha256":"60158d73c7da1bf41ccc2970a9a4b8d45edaf792967b6ffafbda51588b9bfda4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:56:31.842508Z","signature_b64":"QJfB2ZHrJQVDJ/usdtHDaGg3cFalHmp+r8dXU+D6veteMw0tV37t84fHt6zdN1VQlO37yuL28NPdl9P9yIuyDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a8a0a6d1ad4965309fd69188d939a331b984a83c6ac8fb6b37cade408c962a8","last_reissued_at":"2026-07-05T01:56:31.842121Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:56:31.842121Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Graph Neural Networks versus Graph-Augmented MLPs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.CO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Joan Bruna, Lei Chen, Zhengdao Chen","submitted_at":"2020-10-28T17:59:59Z","abstract_excerpt":"From the perspective of expressive power, this work compares multi-layer Graph Neural Networks (GNNs) with a simplified alternative that we call Graph-Augmented Multi-Layer Perceptrons (GA-MLPs), which first augments node features with certain multi-hop operators on the graph and then applies an MLP in a node-wise fashion. From the perspective of graph isomorphism testing, we show both theoretically and numerically that GA-MLPs with suitable operators can distinguish almost all non-isomorphic graphs, just like the Weifeiler-Lehman (WL) test. However, by viewing them as node-level functions and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.15116","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/2010.15116/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":"2010.15116","created_at":"2026-07-05T01:56:31.842176+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.15116v2","created_at":"2026-07-05T01:56:31.842176+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.15116","created_at":"2026-07-05T01:56:31.842176+00:00"},{"alias_kind":"pith_short_12","alias_value":"RKFAU3I22SLF","created_at":"2026-07-05T01:56:31.842176+00:00"},{"alias_kind":"pith_short_16","alias_value":"RKFAU3I22SLFGCP5","created_at":"2026-07-05T01:56:31.842176+00:00"},{"alias_kind":"pith_short_8","alias_value":"RKFAU3I2","created_at":"2026-07-05T01:56:31.842176+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.15405","citing_title":"HOPSE: Scalable Higher-Order Positional and Structural Encoder for Combinatorial Representations","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM","json":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM.json","graph_json":"https://pith.science/api/pith-number/RKFAU3I22SLFGCP5NEMI3E42GM/graph.json","events_json":"https://pith.science/api/pith-number/RKFAU3I22SLFGCP5NEMI3E42GM/events.json","paper":"https://pith.science/paper/RKFAU3I2"},"agent_actions":{"view_html":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM","download_json":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM.json","view_paper":"https://pith.science/paper/RKFAU3I2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.15116&json=true","fetch_graph":"https://pith.science/api/pith-number/RKFAU3I22SLFGCP5NEMI3E42GM/graph.json","fetch_events":"https://pith.science/api/pith-number/RKFAU3I22SLFGCP5NEMI3E42GM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM/action/storage_attestation","attest_author":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM/action/author_attestation","sign_citation":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM/action/citation_signature","submit_replication":"https://pith.science/pith/RKFAU3I22SLFGCP5NEMI3E42GM/action/replication_record"}},"created_at":"2026-07-05T01:56:31.842176+00:00","updated_at":"2026-07-05T01:56:31.842176+00:00"}