{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:HRMCIRROLL5CL2MPW6K4LXRSNN","short_pith_number":"pith:HRMCIRRO","schema_version":"1.0","canonical_sha256":"3c5824462e5afa25e98fb795c5de326b417de26e999a3f058812e2d613755824","source":{"kind":"arxiv","id":"2012.09699","version":2},"attestation_state":"computed","paper":{"title":"A Generalization of Transformer Networks to Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Vijay Prakash Dwivedi, Xavier Bresson","submitted_at":"2020-12-17T16:11:47Z","abstract_excerpt":"We propose a generalization of transformer neural network architecture for arbitrary graphs. The original transformer was designed for Natural Language Processing (NLP), which operates on fully connected graphs representing all connections between the words in a sequence. Such architecture does not leverage the graph connectivity inductive bias, and can perform poorly when the graph topology is important and has not been encoded into the node features. We introduce a graph transformer with four new properties compared to the standard model. First, the attention mechanism is a function of the n"},"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":"2012.09699","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-12-17T16:11:47Z","cross_cats_sorted":[],"title_canon_sha256":"5f389141fc59c2b63377c266dcd92983a469b896b80e54b0c2317edb834e98f2","abstract_canon_sha256":"e91405e07c19862b06c1f39a3165c4a192393f440ca1686c1730003a79332860"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:09:15.532893Z","signature_b64":"xHzJC1RB6X3PE2FyUN4LB5eRUhwYBtkxcoBa081NBbgbiCKOSEmXNphHZ5LM69IavS48ZP78eJBfzvNS9to8Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c5824462e5afa25e98fb795c5de326b417de26e999a3f058812e2d613755824","last_reissued_at":"2026-07-05T02:09:15.532481Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:09:15.532481Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Generalization of Transformer Networks to Graphs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Vijay Prakash Dwivedi, Xavier Bresson","submitted_at":"2020-12-17T16:11:47Z","abstract_excerpt":"We propose a generalization of transformer neural network architecture for arbitrary graphs. The original transformer was designed for Natural Language Processing (NLP), which operates on fully connected graphs representing all connections between the words in a sequence. Such architecture does not leverage the graph connectivity inductive bias, and can perform poorly when the graph topology is important and has not been encoded into the node features. We introduce a graph transformer with four new properties compared to the standard model. First, the attention mechanism is a function of the n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.09699","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/2012.09699/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":"2012.09699","created_at":"2026-07-05T02:09:15.532537+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.09699v2","created_at":"2026-07-05T02:09:15.532537+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.09699","created_at":"2026-07-05T02:09:15.532537+00:00"},{"alias_kind":"pith_short_12","alias_value":"HRMCIRROLL5C","created_at":"2026-07-05T02:09:15.532537+00:00"},{"alias_kind":"pith_short_16","alias_value":"HRMCIRROLL5CL2MP","created_at":"2026-07-05T02:09:15.532537+00:00"},{"alias_kind":"pith_short_8","alias_value":"HRMCIRRO","created_at":"2026-07-05T02:09:15.532537+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":36,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07032","citing_title":"Gauge-Invariant Learnable Spectral Positional Encodings for Directed Graphs via Hermitian Block Krylov Subspaces","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25293","citing_title":"Communicability-Inspired Positional Encoding (CIPE)","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23791","citing_title":"One Generator, Any Process: LLM-Conditioning for the LHC","ref_index":284,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22429","citing_title":"Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation","ref_index":117,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18726","citing_title":"Graph Grounded Cross Attention Transformer Neural Network for Structurally Constrained Full Event Sequence Generation in Predictive Process Monitoring","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17667","citing_title":"Handling Feature Heterogeneity with Learnable Graph Patches","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01621","citing_title":"Spatial Support Matters: Geometry-Aware Graph Fusion for Rainfall Field Reconstruction","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10227","citing_title":"Spatiotemporal Graph Transformer for 3D Neighborhood Interaction and Quality Prediction in Metal Additive Manufacturing","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09898","citing_title":"TRAPS: Therapeutic Response Analysis via Pathway-informed Stratification","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05067","citing_title":"FLAGG: Flexible Autoregressive Graph Generation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14759","citing_title":"Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinement","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14769","citing_title":"Composable Crystals: Controllable Materials Discovery via Concept Learning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15231","citing_title":"Mask-Morph Graph U-Net: A Generalisable Mesh-Based Surrogate for Crashworthiness Field Prediction under Large Geometric Variation","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23791","citing_title":"One Generator, Any Process: LLM-Conditioning for the LHC","ref_index":288,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27144","citing_title":"Is an Image Also Worth 16x16=256 Superpixels? A Framework for Attentional Image Classification","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00304","citing_title":"Modeling Spectral Energy Shifts in Spatio-Temporal Graph Anomaly Detection","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2510.12369","citing_title":"A Hierarchical Quantized Tokenization Framework for Task-Adaptive Graph Representation Learning","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2210.15304","citing_title":"Explaining the Explainers in Graph Neural Networks: a Comparative Study","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22593","citing_title":"Do Deep Ensembles Actually Capture Uncertainty in Graph Neural Networks?","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2603.01388","citing_title":"Invariant-Stratified Propagation for Expressive Graph Neural Networks","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15231","citing_title":"Mask-Morph Graph U-Net: A Generalisable Mesh-Based Surrogate for Crashworthiness Field Prediction under Large Geometric Variation","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15524","citing_title":"Neural Point-Forms","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21000","citing_title":"Feature Augmentation of GNNs for ILPs: Local Uniqueness Suffices","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2510.01801","citing_title":"Detecting LLM-Generated Spam Reviews by Integrating Language Model Embeddings and Graph Neural Network","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2105.14491","citing_title":"How Attentive are Graph Attention Networks?","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN","json":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN.json","graph_json":"https://pith.science/api/pith-number/HRMCIRROLL5CL2MPW6K4LXRSNN/graph.json","events_json":"https://pith.science/api/pith-number/HRMCIRROLL5CL2MPW6K4LXRSNN/events.json","paper":"https://pith.science/paper/HRMCIRRO"},"agent_actions":{"view_html":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN","download_json":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN.json","view_paper":"https://pith.science/paper/HRMCIRRO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.09699&json=true","fetch_graph":"https://pith.science/api/pith-number/HRMCIRROLL5CL2MPW6K4LXRSNN/graph.json","fetch_events":"https://pith.science/api/pith-number/HRMCIRROLL5CL2MPW6K4LXRSNN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN/action/storage_attestation","attest_author":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN/action/author_attestation","sign_citation":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN/action/citation_signature","submit_replication":"https://pith.science/pith/HRMCIRROLL5CL2MPW6K4LXRSNN/action/replication_record"}},"created_at":"2026-07-05T02:09:15.532537+00:00","updated_at":"2026-07-05T02:09:15.532537+00:00"}