{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:E32WNTIBHW2R3IKCJBZQUIIFWE","short_pith_number":"pith:E32WNTIB","schema_version":"1.0","canonical_sha256":"26f566cd013db51da14248730a2105b115dd10101d9993df409d09950995159a","source":{"kind":"arxiv","id":"2308.00261","version":1},"attestation_state":"computed","paper":{"title":"Improving Pixel-based MIM by Reducing Wasted Modeling Capability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiacheng Chen, Kai Chen, Songyang Zhang, Yuan Liu, Zhaohui Yu","submitted_at":"2023-08-01T03:44:56Z","abstract_excerpt":"There has been significant progress in Masked Image Modeling (MIM). Existing MIM methods can be broadly categorized into two groups based on the reconstruction target: pixel-based and tokenizer-based approaches. The former offers a simpler pipeline and lower computational cost, but it is known to be biased toward high-frequency details. In this paper, we provide a set of empirical studies to confirm this limitation of pixel-based MIM and propose a new method that explicitly utilizes low-level features from shallow layers to aid pixel reconstruction. By incorporating this design into our base m"},"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":"2308.00261","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-01T03:44:56Z","cross_cats_sorted":[],"title_canon_sha256":"ec891494f0391d98a155ceb5a8462f02c4895ab821510266876b5618299ccfb4","abstract_canon_sha256":"5df62aaadda6168956b524a7238dd559e7b6c08822df86174e9eca9d06f85daf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:36:16.716446Z","signature_b64":"Yh8ubveJdIIal48ioDkxRtCLMqGICdir4vMvrkkeg4sNKlSY3nykp7lAI77lvo62Js4vJVHNmI2Kpc5c6NB8Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26f566cd013db51da14248730a2105b115dd10101d9993df409d09950995159a","last_reissued_at":"2026-07-05T06:36:16.716014Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:36:16.716014Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Pixel-based MIM by Reducing Wasted Modeling Capability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiacheng Chen, Kai Chen, Songyang Zhang, Yuan Liu, Zhaohui Yu","submitted_at":"2023-08-01T03:44:56Z","abstract_excerpt":"There has been significant progress in Masked Image Modeling (MIM). Existing MIM methods can be broadly categorized into two groups based on the reconstruction target: pixel-based and tokenizer-based approaches. The former offers a simpler pipeline and lower computational cost, but it is known to be biased toward high-frequency details. In this paper, we provide a set of empirical studies to confirm this limitation of pixel-based MIM and propose a new method that explicitly utilizes low-level features from shallow layers to aid pixel reconstruction. By incorporating this design into our base m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.00261","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/2308.00261/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":"2308.00261","created_at":"2026-07-05T06:36:16.716080+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.00261v1","created_at":"2026-07-05T06:36:16.716080+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.00261","created_at":"2026-07-05T06:36:16.716080+00:00"},{"alias_kind":"pith_short_12","alias_value":"E32WNTIBHW2R","created_at":"2026-07-05T06:36:16.716080+00:00"},{"alias_kind":"pith_short_16","alias_value":"E32WNTIBHW2R3IKC","created_at":"2026-07-05T06:36:16.716080+00:00"},{"alias_kind":"pith_short_8","alias_value":"E32WNTIB","created_at":"2026-07-05T06:36:16.716080+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.08819","citing_title":"Thoughts on Objectives of Sparse and Hierarchical Masked Image Model","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE","json":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE.json","graph_json":"https://pith.science/api/pith-number/E32WNTIBHW2R3IKCJBZQUIIFWE/graph.json","events_json":"https://pith.science/api/pith-number/E32WNTIBHW2R3IKCJBZQUIIFWE/events.json","paper":"https://pith.science/paper/E32WNTIB"},"agent_actions":{"view_html":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE","download_json":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE.json","view_paper":"https://pith.science/paper/E32WNTIB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.00261&json=true","fetch_graph":"https://pith.science/api/pith-number/E32WNTIBHW2R3IKCJBZQUIIFWE/graph.json","fetch_events":"https://pith.science/api/pith-number/E32WNTIBHW2R3IKCJBZQUIIFWE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE/action/storage_attestation","attest_author":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE/action/author_attestation","sign_citation":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE/action/citation_signature","submit_replication":"https://pith.science/pith/E32WNTIBHW2R3IKCJBZQUIIFWE/action/replication_record"}},"created_at":"2026-07-05T06:36:16.716080+00:00","updated_at":"2026-07-05T06:36:16.716080+00:00"}