{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4YOXBF4MW6SVA2WURKCCB72J6E","short_pith_number":"pith:4YOXBF4M","schema_version":"1.0","canonical_sha256":"e61d70978cb7a5506ad48a8420ff49f112b19eba50ebb840cbe287547a6d0530","source":{"kind":"arxiv","id":"2110.03336","version":4},"attestation_state":"computed","paper":{"title":"Frame Averaging for Invariant and Equivariant Network Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aditya Grover, Edward J. Smith, Heli Ben-Hamu, Ishan Misra, Matan Atzmon, Omri Puny, Yaron Lipman","submitted_at":"2021-10-07T11:05:23Z","abstract_excerpt":"Many machine learning tasks involve learning functions that are known to be invariant or equivariant to certain symmetries of the input data. However, it is often challenging to design neural network architectures that respect these symmetries while being expressive and computationally efficient. For example, Euclidean motion invariant/equivariant graph or point cloud neural networks. We introduce Frame Averaging (FA), a general purpose and systematic framework for adapting known (backbone) architectures to become invariant or equivariant to new symmetry types. Our framework builds on the well"},"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":"2110.03336","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T11:05:23Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"16bd93b2440621e6f36600c906f7e4270400158a1e9e68341a20a263dc18e208","abstract_canon_sha256":"59133f3e4e76b86d3dd39681781fecea4489f71828212f3bb77f48145e507d59"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:05:35.447363Z","signature_b64":"+Vniu/1OVcm6dVVZBJ9YhXHXUrlsoppBxuhB0v8AYcBl749q1fu73B9Y7afNIbZC2rTuib9tNIlO3xWx62BiBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e61d70978cb7a5506ad48a8420ff49f112b19eba50ebb840cbe287547a6d0530","last_reissued_at":"2026-07-05T04:05:35.446895Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:05:35.446895Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Frame Averaging for Invariant and Equivariant Network Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Aditya Grover, Edward J. Smith, Heli Ben-Hamu, Ishan Misra, Matan Atzmon, Omri Puny, Yaron Lipman","submitted_at":"2021-10-07T11:05:23Z","abstract_excerpt":"Many machine learning tasks involve learning functions that are known to be invariant or equivariant to certain symmetries of the input data. However, it is often challenging to design neural network architectures that respect these symmetries while being expressive and computationally efficient. For example, Euclidean motion invariant/equivariant graph or point cloud neural networks. We introduce Frame Averaging (FA), a general purpose and systematic framework for adapting known (backbone) architectures to become invariant or equivariant to new symmetry types. Our framework builds on the well"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03336","kind":"arxiv","version":4},"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/2110.03336/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":"2110.03336","created_at":"2026-07-05T04:05:35.446956+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.03336v4","created_at":"2026-07-05T04:05:35.446956+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03336","created_at":"2026-07-05T04:05:35.446956+00:00"},{"alias_kind":"pith_short_12","alias_value":"4YOXBF4MW6SV","created_at":"2026-07-05T04:05:35.446956+00:00"},{"alias_kind":"pith_short_16","alias_value":"4YOXBF4MW6SVA2WU","created_at":"2026-07-05T04:05:35.446956+00:00"},{"alias_kind":"pith_short_8","alias_value":"4YOXBF4M","created_at":"2026-07-05T04:05:35.446956+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19784","citing_title":"EquiVLA: A General Framework for Rotationally Equivariant Vision-Language-Action Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09432","citing_title":"Graph Mamba Operator: A Latent Simulator for Interacting Particle Systems","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00556","citing_title":"Group-Equivariant Poincar\\'e Convolutional Networks","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2210.15304","citing_title":"Explaining the Explainers in Graph Neural Networks: a Comparative Study","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24886","citing_title":"Adaptive Canonicalization with Application to Invariant Anisotropic Geometric Networks","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13744","citing_title":"Aligning Network Equivariance with Data Symmetry: A Theoretical Framework and Adaptive Approach for Image Restoration","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11008","citing_title":"When and How to Canonize: A Generalization Perspective","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03520","citing_title":"Parametrizing Convex Sets Using Sublinear Neural Networks","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E","json":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E.json","graph_json":"https://pith.science/api/pith-number/4YOXBF4MW6SVA2WURKCCB72J6E/graph.json","events_json":"https://pith.science/api/pith-number/4YOXBF4MW6SVA2WURKCCB72J6E/events.json","paper":"https://pith.science/paper/4YOXBF4M"},"agent_actions":{"view_html":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E","download_json":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E.json","view_paper":"https://pith.science/paper/4YOXBF4M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.03336&json=true","fetch_graph":"https://pith.science/api/pith-number/4YOXBF4MW6SVA2WURKCCB72J6E/graph.json","fetch_events":"https://pith.science/api/pith-number/4YOXBF4MW6SVA2WURKCCB72J6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E/action/storage_attestation","attest_author":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E/action/author_attestation","sign_citation":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E/action/citation_signature","submit_replication":"https://pith.science/pith/4YOXBF4MW6SVA2WURKCCB72J6E/action/replication_record"}},"created_at":"2026-07-05T04:05:35.446956+00:00","updated_at":"2026-07-05T04:05:35.446956+00:00"}