{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:32GIXJTEZPAQ6NWJYGADIOFZTR","short_pith_number":"pith:32GIXJTE","schema_version":"1.0","canonical_sha256":"de8c8ba664cbc10f36c9c1803438b99c5c4f2c9fabfce287abe35c128a7b0547","source":{"kind":"arxiv","id":"2311.16424","version":1},"attestation_state":"computed","paper":{"title":"Manifold Preserving Guided Diffusion","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chieh-Hsin Lai, Dongjun Kim, J. Zico Kolter, Naoki Murata, Ruslan Salakhutdinov, Stefano Ermon, Toshimitsu Uesaka, Wei-Hsiang Liao, Yuhta Takida, Yuki Mitsufuji, Yutong He","submitted_at":"2023-11-28T02:08:06Z","abstract_excerpt":"Despite the recent advancements, conditional image generation still faces challenges of cost, generalizability, and the need for task-specific training. In this paper, we propose Manifold Preserving Guided Diffusion (MPGD), a training-free conditional generation framework that leverages pretrained diffusion models and off-the-shelf neural networks with minimal additional inference cost for a broad range of tasks. Specifically, we leverage the manifold hypothesis to refine the guided diffusion steps and introduce a shortcut algorithm in the process. We then propose two methods for on-manifold 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":"2311.16424","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-28T02:08:06Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"c519c2402102db1b2901f85eb88be2af6aaa864893c0a4d76484c0cd12c9acdc","abstract_canon_sha256":"347cc8a62bb65bb322805a5e7ccd358b50e96bf1851a1e45e8d8226a4fbb3ea7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:41.483089Z","signature_b64":"bNle1ddwSL2lu07GSyZUoh6QzPn4cfWbx+Ux2PAFfg558v4pVJc1w5Aihq8Df4lQu3pgcerYJEH/sApra4WpBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de8c8ba664cbc10f36c9c1803438b99c5c4f2c9fabfce287abe35c128a7b0547","last_reissued_at":"2026-07-05T07:17:41.482679Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:41.482679Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Manifold Preserving Guided Diffusion","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Chieh-Hsin Lai, Dongjun Kim, J. Zico Kolter, Naoki Murata, Ruslan Salakhutdinov, Stefano Ermon, Toshimitsu Uesaka, Wei-Hsiang Liao, Yuhta Takida, Yuki Mitsufuji, Yutong He","submitted_at":"2023-11-28T02:08:06Z","abstract_excerpt":"Despite the recent advancements, conditional image generation still faces challenges of cost, generalizability, and the need for task-specific training. In this paper, we propose Manifold Preserving Guided Diffusion (MPGD), a training-free conditional generation framework that leverages pretrained diffusion models and off-the-shelf neural networks with minimal additional inference cost for a broad range of tasks. Specifically, we leverage the manifold hypothesis to refine the guided diffusion steps and introduce a shortcut algorithm in the process. We then propose two methods for on-manifold t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.16424","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/2311.16424/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":"2311.16424","created_at":"2026-07-05T07:17:41.482732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.16424v1","created_at":"2026-07-05T07:17:41.482732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.16424","created_at":"2026-07-05T07:17:41.482732+00:00"},{"alias_kind":"pith_short_12","alias_value":"32GIXJTEZPAQ","created_at":"2026-07-05T07:17:41.482732+00:00"},{"alias_kind":"pith_short_16","alias_value":"32GIXJTEZPAQ6NWJ","created_at":"2026-07-05T07:17:41.482732+00:00"},{"alias_kind":"pith_short_8","alias_value":"32GIXJTE","created_at":"2026-07-05T07:17:41.482732+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20404","citing_title":"FlowBender: Feedback-Aware Training for Self-Correcting Conditional Flows","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30370","citing_title":"MUSE: Unlocking Timestep as Native Task Steering for One-Step Dense Prediction","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30991","citing_title":"Parallel Tempering Initial Sampling in Inference-Time Reward Alignment","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07715","citing_title":"Analyzing and Guiding Zero-Shot Posterior Sampling in Diffusion Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11871","citing_title":"$h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2509.25845","citing_title":"Training-Free Reward-Guided Image Editing via Trajectory Optimal Control","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2410.00083","citing_title":"A Survey on Diffusion Models for Inverse Problems","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05285","citing_title":"Robust Inference-Time Steering of Protein Diffusion Models via Embedding Optimization","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11871","citing_title":"$h$-control: Training-Free Camera Control via Block-Conditional Gibbs Refinement","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06779","citing_title":"VASR: Variance-Aware Systematic Resampling for Reward-Guided Diffusion","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03712","citing_title":"Tempered Guided Diffusion","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25289","citing_title":"Exploring Time Conditioning in Diffusion Generative Models from Disjoint Noisy Data Manifolds","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24238","citing_title":"GeoEdit: Local Frames for Fast, Training-Free On-Manifold Editing in Diffusion Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06779","citing_title":"VASR: Variance-Aware Systematic Resampling for Reward-Guided Diffusion","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07661","citing_title":"Stochastic Transition-Map Distillation for Fast Probabilistic Inference","ref_index":147,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR","json":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR.json","graph_json":"https://pith.science/api/pith-number/32GIXJTEZPAQ6NWJYGADIOFZTR/graph.json","events_json":"https://pith.science/api/pith-number/32GIXJTEZPAQ6NWJYGADIOFZTR/events.json","paper":"https://pith.science/paper/32GIXJTE"},"agent_actions":{"view_html":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR","download_json":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR.json","view_paper":"https://pith.science/paper/32GIXJTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.16424&json=true","fetch_graph":"https://pith.science/api/pith-number/32GIXJTEZPAQ6NWJYGADIOFZTR/graph.json","fetch_events":"https://pith.science/api/pith-number/32GIXJTEZPAQ6NWJYGADIOFZTR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR/action/storage_attestation","attest_author":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR/action/author_attestation","sign_citation":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR/action/citation_signature","submit_replication":"https://pith.science/pith/32GIXJTEZPAQ6NWJYGADIOFZTR/action/replication_record"}},"created_at":"2026-07-05T07:17:41.482732+00:00","updated_at":"2026-07-05T07:17:41.482732+00:00"}