{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:I3PMCKTZRUQOOWHX4YJ6SARF3X","short_pith_number":"pith:I3PMCKTZ","schema_version":"1.0","canonical_sha256":"46dec12a798d20e758f7e613e90225ddf75f5f4f1d2218279be93f35f66f8e9e","source":{"kind":"arxiv","id":"2506.20294","version":5},"attestation_state":"computed","paper":{"title":"Ctrl-Z Sampling: Scaling Diffusion Sampling with Controlled Random Zigzag Explorations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chaoyi Zhang, Jieting Long, Ke Xie, Shunqi Mao, Weidong Cai, Wei Guo","submitted_at":"2025-06-25T10:01:00Z","abstract_excerpt":"Diffusion models generate conditional samples by progressively denoising Gaussian noise, yet the denoising trajectory can stall at visually plausible but low-quality outcomes with conditional misalignment or structural artifacts. We interpret this behavior as local optima in a surrogate quality landscape: Once early denoising commits to a suboptimal global structure, later steps mainly sharpen details and seldom correct the underlying mistake. While existing inference-time approaches explore alternative diffusion states via re-noising with fixed strength or direction, they exhibit limited capa"},"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":"2506.20294","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-25T10:01:00Z","cross_cats_sorted":[],"title_canon_sha256":"2f7f7d944b597671bab0bb02daa095096bd872508f0352b166966948f187abce","abstract_canon_sha256":"059d0e59531c99fa863cae12c1f03b3bc755a8d7bc42b142bb96685e6c9ec6d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T01:15:55.551990Z","signature_b64":"ey7ogWHJPyUIB4wF0wa/EZDqksBhelpoFfHXjZaSEUUn/p/GYd5mlfRpbZxgam2UvZ1+Gye/29tJfubWyBIqDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"46dec12a798d20e758f7e613e90225ddf75f5f4f1d2218279be93f35f66f8e9e","last_reissued_at":"2026-07-07T01:15:55.550841Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T01:15:55.550841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ctrl-Z Sampling: Scaling Diffusion Sampling with Controlled Random Zigzag Explorations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chaoyi Zhang, Jieting Long, Ke Xie, Shunqi Mao, Weidong Cai, Wei Guo","submitted_at":"2025-06-25T10:01:00Z","abstract_excerpt":"Diffusion models generate conditional samples by progressively denoising Gaussian noise, yet the denoising trajectory can stall at visually plausible but low-quality outcomes with conditional misalignment or structural artifacts. We interpret this behavior as local optima in a surrogate quality landscape: Once early denoising commits to a suboptimal global structure, later steps mainly sharpen details and seldom correct the underlying mistake. While existing inference-time approaches explore alternative diffusion states via re-noising with fixed strength or direction, they exhibit limited capa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.20294","kind":"arxiv","version":5},"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/2506.20294/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":"2506.20294","created_at":"2026-07-07T01:15:55.550988+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.20294v5","created_at":"2026-07-07T01:15:55.550988+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.20294","created_at":"2026-07-07T01:15:55.550988+00:00"},{"alias_kind":"pith_short_12","alias_value":"I3PMCKTZRUQO","created_at":"2026-07-07T01:15:55.550988+00:00"},{"alias_kind":"pith_short_16","alias_value":"I3PMCKTZRUQOOWHX","created_at":"2026-07-07T01:15:55.550988+00:00"},{"alias_kind":"pith_short_8","alias_value":"I3PMCKTZ","created_at":"2026-07-07T01:15:55.550988+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.19532","citing_title":"Boosting Text-to-Image Diffusion Models via Core Token Attention-Based Seed Selection","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2512.23532","citing_title":"Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X","json":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X.json","graph_json":"https://pith.science/api/pith-number/I3PMCKTZRUQOOWHX4YJ6SARF3X/graph.json","events_json":"https://pith.science/api/pith-number/I3PMCKTZRUQOOWHX4YJ6SARF3X/events.json","paper":"https://pith.science/paper/I3PMCKTZ"},"agent_actions":{"view_html":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X","download_json":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X.json","view_paper":"https://pith.science/paper/I3PMCKTZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.20294&json=true","fetch_graph":"https://pith.science/api/pith-number/I3PMCKTZRUQOOWHX4YJ6SARF3X/graph.json","fetch_events":"https://pith.science/api/pith-number/I3PMCKTZRUQOOWHX4YJ6SARF3X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X/action/storage_attestation","attest_author":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X/action/author_attestation","sign_citation":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X/action/citation_signature","submit_replication":"https://pith.science/pith/I3PMCKTZRUQOOWHX4YJ6SARF3X/action/replication_record"}},"created_at":"2026-07-07T01:15:55.550988+00:00","updated_at":"2026-07-07T01:15:55.550988+00:00"}