{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UA4V25FPMZD4SSAUR7TXF5JH4G","short_pith_number":"pith:UA4V25FP","schema_version":"1.0","canonical_sha256":"a0395d74af6647c948148fe772f527e18794bdf6e78da26de91af7121daab3b6","source":{"kind":"arxiv","id":"2305.18231","version":3},"attestation_state":"computed","paper":{"title":"High-Fidelity Image Compression with Score-based Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG","stat.ML"],"primary_cat":"eess.IV","authors_text":"Eirikur Agustsson, Emiel Hoogeboom, Fabian Mentzer, George Toderici, Lucas Theis, Luca Versari","submitted_at":"2023-05-26T17:16:16Z","abstract_excerpt":"Despite the tremendous success of diffusion generative models in text-to-image generation, replicating this success in the domain of image compression has proven difficult. In this paper, we demonstrate that diffusion can significantly improve perceptual quality at a given bit-rate, outperforming state-of-the-art approaches PO-ELIC and HiFiC as measured by FID score. This is achieved using a simple but theoretically motivated two-stage approach combining an autoencoder targeting MSE followed by a further score-based decoder. However, as we will show, implementation details matter and the optim"},"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":"2305.18231","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.IV","submitted_at":"2023-05-26T17:16:16Z","cross_cats_sorted":["cs.CV","cs.LG","stat.ML"],"title_canon_sha256":"acc7b40255467077e17c6fdbbc739f7f279bb6047872a5cebe38c24b3390dfff","abstract_canon_sha256":"927b9939e96a93f88d93c32c9592a046d9a7e9a96124fcf41a785678510b5420"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:53:30.236235Z","signature_b64":"MMQgPwGQ05zHT9lhx3a4zh8R2+gh1oyc8r3Zn/VdhR5yNpPYiVu10sW9u9SBewU8osTRR2a1QQ+vUaWQkRS3Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0395d74af6647c948148fe772f527e18794bdf6e78da26de91af7121daab3b6","last_reissued_at":"2026-07-05T07:53:30.235767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:53:30.235767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"High-Fidelity Image Compression with Score-based Generative Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.LG","stat.ML"],"primary_cat":"eess.IV","authors_text":"Eirikur Agustsson, Emiel Hoogeboom, Fabian Mentzer, George Toderici, Lucas Theis, Luca Versari","submitted_at":"2023-05-26T17:16:16Z","abstract_excerpt":"Despite the tremendous success of diffusion generative models in text-to-image generation, replicating this success in the domain of image compression has proven difficult. In this paper, we demonstrate that diffusion can significantly improve perceptual quality at a given bit-rate, outperforming state-of-the-art approaches PO-ELIC and HiFiC as measured by FID score. This is achieved using a simple but theoretically motivated two-stage approach combining an autoencoder targeting MSE followed by a further score-based decoder. However, as we will show, implementation details matter and the optim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18231","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/2305.18231/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":"2305.18231","created_at":"2026-07-05T07:53:30.235823+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.18231v3","created_at":"2026-07-05T07:53:30.235823+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18231","created_at":"2026-07-05T07:53:30.235823+00:00"},{"alias_kind":"pith_short_12","alias_value":"UA4V25FPMZD4","created_at":"2026-07-05T07:53:30.235823+00:00"},{"alias_kind":"pith_short_16","alias_value":"UA4V25FPMZD4SSAU","created_at":"2026-07-05T07:53:30.235823+00:00"},{"alias_kind":"pith_short_8","alias_value":"UA4V25FP","created_at":"2026-07-05T07:53:30.235823+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10450","citing_title":"Few-step Generative Models as Lossy Compression","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26571","citing_title":"GVCC: Zero-Shot Video Compression via Codebook-Driven Stochastic Rectified Flow","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10546","citing_title":"Differentiable Vector Quantization for Rate-Distortion Optimization of Generative Image Compression","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06568","citing_title":"A Noise Constrained Diffusion (NC-Diffusion) Framework for High Fidelity Image Compression","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05743","citing_title":"On the Robustness of Diffusion-Based Image Compression to Bit-Flip Errors","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G","json":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G.json","graph_json":"https://pith.science/api/pith-number/UA4V25FPMZD4SSAUR7TXF5JH4G/graph.json","events_json":"https://pith.science/api/pith-number/UA4V25FPMZD4SSAUR7TXF5JH4G/events.json","paper":"https://pith.science/paper/UA4V25FP"},"agent_actions":{"view_html":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G","download_json":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G.json","view_paper":"https://pith.science/paper/UA4V25FP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.18231&json=true","fetch_graph":"https://pith.science/api/pith-number/UA4V25FPMZD4SSAUR7TXF5JH4G/graph.json","fetch_events":"https://pith.science/api/pith-number/UA4V25FPMZD4SSAUR7TXF5JH4G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G/action/storage_attestation","attest_author":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G/action/author_attestation","sign_citation":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G/action/citation_signature","submit_replication":"https://pith.science/pith/UA4V25FPMZD4SSAUR7TXF5JH4G/action/replication_record"}},"created_at":"2026-07-05T07:53:30.235823+00:00","updated_at":"2026-07-05T07:53:30.235823+00:00"}