{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:O4INM62SANJ44TDXAGYUI4KC2B","short_pith_number":"pith:O4INM62S","schema_version":"1.0","canonical_sha256":"7710d67b520353ce4c7701b1447142d058d5747e127a3769988fa9af32b0546e","source":{"kind":"arxiv","id":"2304.07735","version":3},"attestation_state":"computed","paper":{"title":"Permutation Equivariance of Transformers and Its Applications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Baochun Li, Dixi Yao, Hangyu Ye, Hengyuan Xu, Liyao Xiang, Pengzhi Chu","submitted_at":"2023-04-16T09:25:24Z","abstract_excerpt":"Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these models are robust to shuffling but are limited to inter-token permutation in the forward propagation. In this work, we propose our definition of permutation equivariance, a broader concept covering both inter- and intra- token permutation in the forward and backward propagation of neural networks. We rigorously proved that such permutation equivariance property can be satisfied on most vanilla Transformer-based models with almost no adapt"},"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":"2304.07735","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-04-16T09:25:24Z","cross_cats_sorted":[],"title_canon_sha256":"44ace5369112464ed74ff75f052ccb4324e62986725ffdf11af683a78e073885","abstract_canon_sha256":"fab7e53e49f72939f9758e7ee30a095724d0e2617fcef9b2928705a2d6fc45a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:43.025688Z","signature_b64":"ouuZkGxDxCMIpFBgH8Xzj1dO1NR+CfSjQaU4tCRUcwnZXbCZlk7xK07OlgNp+EdHBC62TM190jeR3VQj+zAUCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7710d67b520353ce4c7701b1447142d058d5747e127a3769988fa9af32b0546e","last_reissued_at":"2026-07-05T08:02:43.025220Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:43.025220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Permutation Equivariance of Transformers and Its Applications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Baochun Li, Dixi Yao, Hangyu Ye, Hengyuan Xu, Liyao Xiang, Pengzhi Chu","submitted_at":"2023-04-16T09:25:24Z","abstract_excerpt":"Revolutionizing the field of deep learning, Transformer-based models have achieved remarkable performance in many tasks. Recent research has recognized these models are robust to shuffling but are limited to inter-token permutation in the forward propagation. In this work, we propose our definition of permutation equivariance, a broader concept covering both inter- and intra- token permutation in the forward and backward propagation of neural networks. We rigorously proved that such permutation equivariance property can be satisfied on most vanilla Transformer-based models with almost no adapt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.07735","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/2304.07735/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":"2304.07735","created_at":"2026-07-05T08:02:43.025274+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.07735v3","created_at":"2026-07-05T08:02:43.025274+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.07735","created_at":"2026-07-05T08:02:43.025274+00:00"},{"alias_kind":"pith_short_12","alias_value":"O4INM62SANJ4","created_at":"2026-07-05T08:02:43.025274+00:00"},{"alias_kind":"pith_short_16","alias_value":"O4INM62SANJ44TDX","created_at":"2026-07-05T08:02:43.025274+00:00"},{"alias_kind":"pith_short_8","alias_value":"O4INM62S","created_at":"2026-07-05T08:02:43.025274+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19594","citing_title":"Unsupervised Causal Abstractions Discovery","ref_index":75,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B","json":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B.json","graph_json":"https://pith.science/api/pith-number/O4INM62SANJ44TDXAGYUI4KC2B/graph.json","events_json":"https://pith.science/api/pith-number/O4INM62SANJ44TDXAGYUI4KC2B/events.json","paper":"https://pith.science/paper/O4INM62S"},"agent_actions":{"view_html":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B","download_json":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B.json","view_paper":"https://pith.science/paper/O4INM62S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.07735&json=true","fetch_graph":"https://pith.science/api/pith-number/O4INM62SANJ44TDXAGYUI4KC2B/graph.json","fetch_events":"https://pith.science/api/pith-number/O4INM62SANJ44TDXAGYUI4KC2B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B/action/storage_attestation","attest_author":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B/action/author_attestation","sign_citation":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B/action/citation_signature","submit_replication":"https://pith.science/pith/O4INM62SANJ44TDXAGYUI4KC2B/action/replication_record"}},"created_at":"2026-07-05T08:02:43.025274+00:00","updated_at":"2026-07-05T08:02:43.025274+00:00"}