{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2LKGXBTWAXE5ROONEFOGRZ5FFG","short_pith_number":"pith:2LKGXBTW","schema_version":"1.0","canonical_sha256":"d2d46b867605c9d8b9cd215c68e7a52985a73892135e908bac7020dbe6ac2039","source":{"kind":"arxiv","id":"2412.03895","version":1},"attestation_state":"computed","paper":{"title":"A Noise is Worth Diffusion Guidance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Donghoon Ahn, Eunju Cha, Hyoungwon Cho, Jaewon Min, Jiwon Kang, Kyong Hwan Jin, Minjae Kim, Sanghyun Lee, Sayak Paul, Seonhwa Kim, Seungryong Kim, Wooseok Jang","submitted_at":"2024-12-05T06:09:56Z","abstract_excerpt":"Diffusion models excel in generating high-quality images. However, current diffusion models struggle to produce reliable images without guidance methods, such as classifier-free guidance (CFG). Are guidance methods truly necessary? Observing that noise obtained via diffusion inversion can reconstruct high-quality images without guidance, we focus on the initial noise of the denoising pipeline. By mapping Gaussian noise to `guidance-free noise', we uncover that small low-magnitude low-frequency components significantly enhance the denoising process, removing the need for guidance and thus impro"},"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":"2412.03895","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-05T06:09:56Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"8a0e6722c5bca8999a2acd3f99c9e6a50282561484afff409cdeb8aa38243640","abstract_canon_sha256":"601878604fc8feba6674982cce9d843bea4d19d13a828c14235fa2548ca80b5d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:44:56.012993Z","signature_b64":"mBRQy0w5U9E9yn9AJQtKsVjHIvrqS4ot/oHBwsJocEpURfhJXWBHWRpHpI048UhV+XwBCdph3YpCchtU8/QmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2d46b867605c9d8b9cd215c68e7a52985a73892135e908bac7020dbe6ac2039","last_reissued_at":"2026-07-05T09:44:56.012523Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:44:56.012523Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Noise is Worth Diffusion Guidance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Donghoon Ahn, Eunju Cha, Hyoungwon Cho, Jaewon Min, Jiwon Kang, Kyong Hwan Jin, Minjae Kim, Sanghyun Lee, Sayak Paul, Seonhwa Kim, Seungryong Kim, Wooseok Jang","submitted_at":"2024-12-05T06:09:56Z","abstract_excerpt":"Diffusion models excel in generating high-quality images. However, current diffusion models struggle to produce reliable images without guidance methods, such as classifier-free guidance (CFG). Are guidance methods truly necessary? Observing that noise obtained via diffusion inversion can reconstruct high-quality images without guidance, we focus on the initial noise of the denoising pipeline. By mapping Gaussian noise to `guidance-free noise', we uncover that small low-magnitude low-frequency components significantly enhance the denoising process, removing the need for guidance and thus impro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03895","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/2412.03895/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":"2412.03895","created_at":"2026-07-05T09:44:56.012580+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.03895v1","created_at":"2026-07-05T09:44:56.012580+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03895","created_at":"2026-07-05T09:44:56.012580+00:00"},{"alias_kind":"pith_short_12","alias_value":"2LKGXBTWAXE5","created_at":"2026-07-05T09:44:56.012580+00:00"},{"alias_kind":"pith_short_16","alias_value":"2LKGXBTWAXE5ROON","created_at":"2026-07-05T09:44:56.012580+00:00"},{"alias_kind":"pith_short_8","alias_value":"2LKGXBTW","created_at":"2026-07-05T09:44:56.012580+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17408","citing_title":"Where Should Action Generation Begin? A Learnable Source Prior for Generative Robot Policies","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30332","citing_title":"Colored Noise Diffusion Sampling","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30991","citing_title":"Parallel Tempering Initial Sampling in Inference-Time Reward Alignment","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2501.09732","citing_title":"Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2507.06121","citing_title":"Brownian Bridge Diffusion for Sequential Recommendation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2506.15799","citing_title":"Steering Your Diffusion Policy with Latent Space Reinforcement Learning","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00090","citing_title":"It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07322","citing_title":"Action-to-Action Flow Matching","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08144","citing_title":"NoiseRater: Meta-Learned Noise Valuation for Diffusion Model Training","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23540","citing_title":"Oracle Noise: Faster Semantic Spherical Alignment for Interpretable Latent Optimization","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19730","citing_title":"FASTER: Value-Guided Sampling for Fast RL","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG","json":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG.json","graph_json":"https://pith.science/api/pith-number/2LKGXBTWAXE5ROONEFOGRZ5FFG/graph.json","events_json":"https://pith.science/api/pith-number/2LKGXBTWAXE5ROONEFOGRZ5FFG/events.json","paper":"https://pith.science/paper/2LKGXBTW"},"agent_actions":{"view_html":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG","download_json":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG.json","view_paper":"https://pith.science/paper/2LKGXBTW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.03895&json=true","fetch_graph":"https://pith.science/api/pith-number/2LKGXBTWAXE5ROONEFOGRZ5FFG/graph.json","fetch_events":"https://pith.science/api/pith-number/2LKGXBTWAXE5ROONEFOGRZ5FFG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG/action/storage_attestation","attest_author":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG/action/author_attestation","sign_citation":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG/action/citation_signature","submit_replication":"https://pith.science/pith/2LKGXBTWAXE5ROONEFOGRZ5FFG/action/replication_record"}},"created_at":"2026-07-05T09:44:56.012580+00:00","updated_at":"2026-07-05T09:44:56.012580+00:00"}