{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AJWBQZFE5VWZ7B4SEMFWRGJJVD","short_pith_number":"pith:AJWBQZFE","schema_version":"1.0","canonical_sha256":"026c1864a4ed6d9f8792230b689929a8c0fbe57963d2e902fe11d9a28a01b255","source":{"kind":"arxiv","id":"2401.01808","version":1},"attestation_state":"computed","paper":{"title":"aMUSEd: An Open MUSE Reproduction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Patrick von Platen, Robin Rombach, Suraj Patil, William Berman","submitted_at":"2024-01-03T16:10:07Z","abstract_excerpt":"We present aMUSEd, an open-source, lightweight masked image model (MIM) for text-to-image generation based on MUSE. With 10 percent of MUSE's parameters, aMUSEd is focused on fast image generation. We believe MIM is under-explored compared to latent diffusion, the prevailing approach for text-to-image generation. Compared to latent diffusion, MIM requires fewer inference steps and is more interpretable. Additionally, MIM can be fine-tuned to learn additional styles with only a single image. We hope to encourage further exploration of MIM by demonstrating its effectiveness on large-scale text-t"},"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":"2401.01808","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-03T16:10:07Z","cross_cats_sorted":[],"title_canon_sha256":"abaf951a1c12bb5bfa29dabe7ebed86f7922457775e3e6e9c09fb014c7cd9c4b","abstract_canon_sha256":"aeecfab5fac3b58c1ad9381d18a45fe458e3470f6a9bad73a904f02be3fd11cf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:29:57.643768Z","signature_b64":"xVOb3WyJrsISPWtOAtvThKEFjM9yA2DWhspJzZecXqZnRvcJFF2kVPHW5KMeGhk8OvWQts83YypNqE9QFlOCCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"026c1864a4ed6d9f8792230b689929a8c0fbe57963d2e902fe11d9a28a01b255","last_reissued_at":"2026-07-05T07:29:57.643288Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:29:57.643288Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"aMUSEd: An Open MUSE Reproduction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Patrick von Platen, Robin Rombach, Suraj Patil, William Berman","submitted_at":"2024-01-03T16:10:07Z","abstract_excerpt":"We present aMUSEd, an open-source, lightweight masked image model (MIM) for text-to-image generation based on MUSE. With 10 percent of MUSE's parameters, aMUSEd is focused on fast image generation. We believe MIM is under-explored compared to latent diffusion, the prevailing approach for text-to-image generation. Compared to latent diffusion, MIM requires fewer inference steps and is more interpretable. Additionally, MIM can be fine-tuned to learn additional styles with only a single image. We hope to encourage further exploration of MIM by demonstrating its effectiveness on large-scale text-t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.01808","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/2401.01808/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":"2401.01808","created_at":"2026-07-05T07:29:57.643355+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.01808v1","created_at":"2026-07-05T07:29:57.643355+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.01808","created_at":"2026-07-05T07:29:57.643355+00:00"},{"alias_kind":"pith_short_12","alias_value":"AJWBQZFE5VWZ","created_at":"2026-07-05T07:29:57.643355+00:00"},{"alias_kind":"pith_short_16","alias_value":"AJWBQZFE5VWZ7B4S","created_at":"2026-07-05T07:29:57.643355+00:00"},{"alias_kind":"pith_short_8","alias_value":"AJWBQZFE","created_at":"2026-07-05T07:29:57.643355+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10309","citing_title":"Dissect and Prune: Enhancing Robustness in AI-Generated Image Detection","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2405.06535","citing_title":"Controllable Image Generation with Composed Parallel Token Prediction","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2512.19433","citing_title":"dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language Models","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2402.08268","citing_title":"World Model on Million-Length Video And Language With Blockwise RingAttention","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05730","citing_title":"Controllable Image Generation with Composed Parallel Token Prediction","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD","json":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD.json","graph_json":"https://pith.science/api/pith-number/AJWBQZFE5VWZ7B4SEMFWRGJJVD/graph.json","events_json":"https://pith.science/api/pith-number/AJWBQZFE5VWZ7B4SEMFWRGJJVD/events.json","paper":"https://pith.science/paper/AJWBQZFE"},"agent_actions":{"view_html":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD","download_json":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD.json","view_paper":"https://pith.science/paper/AJWBQZFE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.01808&json=true","fetch_graph":"https://pith.science/api/pith-number/AJWBQZFE5VWZ7B4SEMFWRGJJVD/graph.json","fetch_events":"https://pith.science/api/pith-number/AJWBQZFE5VWZ7B4SEMFWRGJJVD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD/action/storage_attestation","attest_author":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD/action/author_attestation","sign_citation":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD/action/citation_signature","submit_replication":"https://pith.science/pith/AJWBQZFE5VWZ7B4SEMFWRGJJVD/action/replication_record"}},"created_at":"2026-07-05T07:29:57.643355+00:00","updated_at":"2026-07-05T07:29:57.643355+00:00"}