{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SVWHTSU35ATXBPUXTPUUTKOUAO","short_pith_number":"pith:SVWHTSU3","schema_version":"1.0","canonical_sha256":"956c79ca9be82770be979be949a9d40381f5a5819c811319edc3a232c77d517c","source":{"kind":"arxiv","id":"2203.14415","version":1},"attestation_state":"computed","paper":{"title":"Mugs: A Multi-Granular Self-Supervised Learning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chenyang Si, Pan Zhou, Shuicheng Yan, Teck Khim Ng, Weihao Yu, Yichen Zhou","submitted_at":"2022-03-27T23:42:05Z","abstract_excerpt":"In self-supervised learning, multi-granular features are heavily desired though rarely investigated, as different downstream tasks (e.g., general and fine-grained classification) often require different or multi-granular features, e.g.~fine- or coarse-grained one or their mixture. In this work, for the first time, we propose an effective MUlti-Granular Self-supervised learning (Mugs) framework to explicitly learn multi-granular visual features. Mugs has three complementary granular supervisions: 1) an instance discrimination supervision (IDS), 2) a novel local-group discrimination supervision "},"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":"2203.14415","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-27T23:42:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9667958a24fce05c02dff7d8c05a6a0c43f83aecf796a87d700169a6e62169d4","abstract_canon_sha256":"1965f8848ba69b2a19472cbc9ce04e0133e73fcd25d3ee8657419654e446c724"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:09:04.700383Z","signature_b64":"FhDmQ5uPyqoaxreYPyTE3mxa04l6tGDmz1nYWfrpEXZGVst6ti3Ty3FtkoIitkcLm39bylbR4qcJO4RlhnS8AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"956c79ca9be82770be979be949a9d40381f5a5819c811319edc3a232c77d517c","last_reissued_at":"2026-07-05T04:09:04.699961Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:09:04.699961Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mugs: A Multi-Granular Self-Supervised Learning Framework","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chenyang Si, Pan Zhou, Shuicheng Yan, Teck Khim Ng, Weihao Yu, Yichen Zhou","submitted_at":"2022-03-27T23:42:05Z","abstract_excerpt":"In self-supervised learning, multi-granular features are heavily desired though rarely investigated, as different downstream tasks (e.g., general and fine-grained classification) often require different or multi-granular features, e.g.~fine- or coarse-grained one or their mixture. In this work, for the first time, we propose an effective MUlti-Granular Self-supervised learning (Mugs) framework to explicitly learn multi-granular visual features. Mugs has three complementary granular supervisions: 1) an instance discrimination supervision (IDS), 2) a novel local-group discrimination supervision "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.14415","kind":"arxiv","version":1},"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/2203.14415/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":"2203.14415","created_at":"2026-07-05T04:09:04.700015+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.14415v1","created_at":"2026-07-05T04:09:04.700015+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.14415","created_at":"2026-07-05T04:09:04.700015+00:00"},{"alias_kind":"pith_short_12","alias_value":"SVWHTSU35ATX","created_at":"2026-07-05T04:09:04.700015+00:00"},{"alias_kind":"pith_short_16","alias_value":"SVWHTSU35ATXBPUX","created_at":"2026-07-05T04:09:04.700015+00:00"},{"alias_kind":"pith_short_8","alias_value":"SVWHTSU3","created_at":"2026-07-05T04:09:04.700015+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.29691","citing_title":"Unsupervised Semantic Segmentation Facilitates Model Understanding","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29691","citing_title":"Unsupervised Semantic Segmentation Facilitates Model Understanding","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14137","citing_title":"Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2304.07193","citing_title":"DINOv2: Learning Robust Visual Features without Supervision","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO","json":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO.json","graph_json":"https://pith.science/api/pith-number/SVWHTSU35ATXBPUXTPUUTKOUAO/graph.json","events_json":"https://pith.science/api/pith-number/SVWHTSU35ATXBPUXTPUUTKOUAO/events.json","paper":"https://pith.science/paper/SVWHTSU3"},"agent_actions":{"view_html":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO","download_json":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO.json","view_paper":"https://pith.science/paper/SVWHTSU3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.14415&json=true","fetch_graph":"https://pith.science/api/pith-number/SVWHTSU35ATXBPUXTPUUTKOUAO/graph.json","fetch_events":"https://pith.science/api/pith-number/SVWHTSU35ATXBPUXTPUUTKOUAO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO/action/storage_attestation","attest_author":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO/action/author_attestation","sign_citation":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO/action/citation_signature","submit_replication":"https://pith.science/pith/SVWHTSU35ATXBPUXTPUUTKOUAO/action/replication_record"}},"created_at":"2026-07-05T04:09:04.700015+00:00","updated_at":"2026-07-05T04:09:04.700015+00:00"}