{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:5HHPNCLSGRV3LPRNMPEWWKC26B","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"01dc9a200176e04f4fa8bca13d9537767c37c8805565af2138e64257041f3f4b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-07T15:24:54Z","title_canon_sha256":"1aba86087772450a1845c38ee919c59c18f1953faec2daffc4001ed7055096be"},"schema_version":"1.0","source":{"id":"2106.03714","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.03714","created_at":"2026-07-05T02:46:52Z"},{"alias_kind":"arxiv_version","alias_value":"2106.03714v1","created_at":"2026-07-05T02:46:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.03714","created_at":"2026-07-05T02:46:52Z"},{"alias_kind":"pith_short_12","alias_value":"5HHPNCLSGRV3","created_at":"2026-07-05T02:46:52Z"},{"alias_kind":"pith_short_16","alias_value":"5HHPNCLSGRV3LPRN","created_at":"2026-07-05T02:46:52Z"},{"alias_kind":"pith_short_8","alias_value":"5HHPNCLS","created_at":"2026-07-05T02:46:52Z"}],"graph_snapshots":[{"event_id":"sha256:8c98dfe50c43144f295b9bdbdd512379e9b42ebb058f96df9b5aa1b2eb5be951","target":"graph","created_at":"2026-07-05T02:46:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2106.03714/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Vision Transformers (ViTs) have shown competitive accuracy in image classification tasks compared with CNNs. Yet, they generally require much more data for model pre-training. Most of recent works thus are dedicated to designing more complex architectures or training methods to address the data-efficiency issue of ViTs. However, few of them explore improving the self-attention mechanism, a key factor distinguishing ViTs from CNNs. Different from existing works, we introduce a conceptually simple scheme, called refiner, to directly refine the self-attention maps of ViTs. Specifically, refiner e","authors_text":"Bingyi Kang, Daquan Zhou, Jiashi Feng, Qibin Hou, Weihao Yu, Xiaojie Jin, Yuan Li, Yujun Shi, Zihang Jiang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-07T15:24:54Z","title":"Refiner: Refining Self-attention for Vision Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.03714","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3388a1ced0f5df20cb518ae5788ac1cdd2676957c9b028c2e960a4259b432331","target":"record","created_at":"2026-07-05T02:46:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"01dc9a200176e04f4fa8bca13d9537767c37c8805565af2138e64257041f3f4b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-07T15:24:54Z","title_canon_sha256":"1aba86087772450a1845c38ee919c59c18f1953faec2daffc4001ed7055096be"},"schema_version":"1.0","source":{"id":"2106.03714","kind":"arxiv","version":1}},"canonical_sha256":"e9cef68972346bb5be2d63c96b285af0508db94fe0d39e9776710d1a6b7dca53","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e9cef68972346bb5be2d63c96b285af0508db94fe0d39e9776710d1a6b7dca53","first_computed_at":"2026-07-05T02:46:52.450850Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:46:52.450850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rNxBxOcizvLFf94qMsTHK8er1jtPoSuF1VWs9k/Q6aSvH3YK0v+/jTFywQUo0vwONRJBHo7P2ZQy7R43Hrx+Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:46:52.451339Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.03714","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3388a1ced0f5df20cb518ae5788ac1cdd2676957c9b028c2e960a4259b432331","sha256:8c98dfe50c43144f295b9bdbdd512379e9b42ebb058f96df9b5aa1b2eb5be951"],"state_sha256":"629af85898f4178b31092fe46d02838ec6589885a83af2f03f492d9d96b7322c"}