{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FAMJBYC4N7SPJOCNRN3FQ4ABOW","short_pith_number":"pith:FAMJBYC4","schema_version":"1.0","canonical_sha256":"281890e05c6fe4f4b84d8b7658700175b25d5a3c1152a889d3820f119bc04197","source":{"kind":"arxiv","id":"2305.18415","version":3},"attestation_state":"computed","paper":{"title":"Geometric Algebra Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Johann Brehmer, Pim de Haan, S\\\"onke Behrends, Taco Cohen","submitted_at":"2023-05-28T18:48:50Z","abstract_excerpt":"Problems involving geometric data arise in physics, chemistry, robotics, computer vision, and many other fields. Such data can take numerous forms, for instance points, direction vectors, translations, or rotations, but to date there is no single architecture that can be applied to such a wide variety of geometric types while respecting their symmetries. In this paper we introduce the Geometric Algebra Transformer (GATr), a general-purpose architecture for geometric data. GATr represents inputs, outputs, and hidden states in the projective geometric (or Clifford) algebra, which offers an effic"},"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":"2305.18415","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-05-28T18:48:50Z","cross_cats_sorted":["cs.RO","stat.ML"],"title_canon_sha256":"c6b7375e950856ee88b12bee8d27a213753ffd91e61030737d476edc0f59a841","abstract_canon_sha256":"9cc4e1993b7b31250ba928880d9bfc027583cd298991a865177d2869783b0ad5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:14:12.897943Z","signature_b64":"RnHkQxAJEkCaTO+Wg8lX4bPzoGu7cy6bgUdtMGbFUNwPFaMUYssFB5XEDHjBD7Rp9u5Q35Pxo8YvJaitvFuNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"281890e05c6fe4f4b84d8b7658700175b25d5a3c1152a889d3820f119bc04197","last_reissued_at":"2026-07-05T07:14:12.897383Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:14:12.897383Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Geometric Algebra Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Johann Brehmer, Pim de Haan, S\\\"onke Behrends, Taco Cohen","submitted_at":"2023-05-28T18:48:50Z","abstract_excerpt":"Problems involving geometric data arise in physics, chemistry, robotics, computer vision, and many other fields. Such data can take numerous forms, for instance points, direction vectors, translations, or rotations, but to date there is no single architecture that can be applied to such a wide variety of geometric types while respecting their symmetries. In this paper we introduce the Geometric Algebra Transformer (GATr), a general-purpose architecture for geometric data. GATr represents inputs, outputs, and hidden states in the projective geometric (or Clifford) algebra, which offers an effic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18415","kind":"arxiv","version":3},"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/2305.18415/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":"2305.18415","created_at":"2026-07-05T07:14:12.897506+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18415v3","created_at":"2026-07-05T07:14:12.897506+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18415","created_at":"2026-07-05T07:14:12.897506+00:00"},{"alias_kind":"pith_short_12","alias_value":"FAMJBYC4N7SP","created_at":"2026-07-05T07:14:12.897506+00:00"},{"alias_kind":"pith_short_16","alias_value":"FAMJBYC4N7SPJOCN","created_at":"2026-07-05T07:14:12.897506+00:00"},{"alias_kind":"pith_short_8","alias_value":"FAMJBYC4","created_at":"2026-07-05T07:14:12.897506+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21790","citing_title":"What Do Lorentz-Equivariant Jet Taggers Learn?","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22891","citing_title":"Pointwise Metrics Mislead: An Evaluation Protocol for Multimodal Inverse Problems","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21483","citing_title":"Velocityformer: Broken-Symmetry-Matched Equivariant Graph Transformers for Cosmological Velocity Reconstruction","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15910","citing_title":"Geometric algebra as the input language of collider foundation models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2510.05482","citing_title":"ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamics","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW","json":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW.json","graph_json":"https://pith.science/api/pith-number/FAMJBYC4N7SPJOCNRN3FQ4ABOW/graph.json","events_json":"https://pith.science/api/pith-number/FAMJBYC4N7SPJOCNRN3FQ4ABOW/events.json","paper":"https://pith.science/paper/FAMJBYC4"},"agent_actions":{"view_html":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW","download_json":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW.json","view_paper":"https://pith.science/paper/FAMJBYC4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18415&json=true","fetch_graph":"https://pith.science/api/pith-number/FAMJBYC4N7SPJOCNRN3FQ4ABOW/graph.json","fetch_events":"https://pith.science/api/pith-number/FAMJBYC4N7SPJOCNRN3FQ4ABOW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW/action/storage_attestation","attest_author":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW/action/author_attestation","sign_citation":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW/action/citation_signature","submit_replication":"https://pith.science/pith/FAMJBYC4N7SPJOCNRN3FQ4ABOW/action/replication_record"}},"created_at":"2026-07-05T07:14:12.897506+00:00","updated_at":"2026-07-05T07:14:12.897506+00:00"}