{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:BPEYYDR2PZT2KDHN5S2T4YRMEM","short_pith_number":"pith:BPEYYDR2","schema_version":"1.0","canonical_sha256":"0bc98c0e3a7e67a50cedecb53e622c2335c83651d02a571423f65f564b439f2b","source":{"kind":"arxiv","id":"2508.12361","version":1},"attestation_state":"computed","paper":{"title":"Navigating the Exploration-Exploitation Tradeoff in Inference-Time Scaling of Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Hiroyuki Kasai, Jianming Huang, Xun Su, Yang Yusen, Zhongxi Fang","submitted_at":"2025-08-17T13:35:38Z","abstract_excerpt":"Inference-time scaling has achieved remarkable success in language models, yet its adaptation to diffusion models remains underexplored. We observe that the efficacy of recent Sequential Monte Carlo (SMC)-based methods largely stems from globally fitting the The reward-tilted distribution, which inherently preserves diversity during multi-modal search. However, current applications of SMC to diffusion models face a fundamental dilemma: early-stage noise samples offer high potential for improvement but are difficult to evaluate accurately, whereas late-stage samples can be reliably assessed but"},"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":"2508.12361","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-17T13:35:38Z","cross_cats_sorted":["cs.AI","math.ST","stat.TH"],"title_canon_sha256":"f94a0cbb2c858fa4fdb1f2a3a6b7af405db3530936847f55e9bd4980ce1bd64e","abstract_canon_sha256":"5736ca427612f689fc232fa16bc05e8c1a60093bf94aeb77ffad5b2016d9cb3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:04.333699Z","signature_b64":"o+B7T5mg824BYNGqDX0MDfWmPCW/idUgRBei6NotpgKRsxNAoeXLgpaqiJWSRLjCI0bShJIqMKuL7GHO05EzBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0bc98c0e3a7e67a50cedecb53e622c2335c83651d02a571423f65f564b439f2b","last_reissued_at":"2026-07-05T11:55:04.333220Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:04.333220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Navigating the Exploration-Exploitation Tradeoff in Inference-Time Scaling of Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","math.ST","stat.TH"],"primary_cat":"cs.LG","authors_text":"Hiroyuki Kasai, Jianming Huang, Xun Su, Yang Yusen, Zhongxi Fang","submitted_at":"2025-08-17T13:35:38Z","abstract_excerpt":"Inference-time scaling has achieved remarkable success in language models, yet its adaptation to diffusion models remains underexplored. We observe that the efficacy of recent Sequential Monte Carlo (SMC)-based methods largely stems from globally fitting the The reward-tilted distribution, which inherently preserves diversity during multi-modal search. However, current applications of SMC to diffusion models face a fundamental dilemma: early-stage noise samples offer high potential for improvement but are difficult to evaluate accurately, whereas late-stage samples can be reliably assessed but"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.12361","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/2508.12361/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":"2508.12361","created_at":"2026-07-05T11:55:04.333278+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.12361v1","created_at":"2026-07-05T11:55:04.333278+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.12361","created_at":"2026-07-05T11:55:04.333278+00:00"},{"alias_kind":"pith_short_12","alias_value":"BPEYYDR2PZT2","created_at":"2026-07-05T11:55:04.333278+00:00"},{"alias_kind":"pith_short_16","alias_value":"BPEYYDR2PZT2KDHN","created_at":"2026-07-05T11:55:04.333278+00:00"},{"alias_kind":"pith_short_8","alias_value":"BPEYYDR2","created_at":"2026-07-05T11:55:04.333278+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2512.23532","citing_title":"Iterative Inference-time Scaling with Adaptive Frequency Steering for Image Super-Resolution","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM","json":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM.json","graph_json":"https://pith.science/api/pith-number/BPEYYDR2PZT2KDHN5S2T4YRMEM/graph.json","events_json":"https://pith.science/api/pith-number/BPEYYDR2PZT2KDHN5S2T4YRMEM/events.json","paper":"https://pith.science/paper/BPEYYDR2"},"agent_actions":{"view_html":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM","download_json":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM.json","view_paper":"https://pith.science/paper/BPEYYDR2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.12361&json=true","fetch_graph":"https://pith.science/api/pith-number/BPEYYDR2PZT2KDHN5S2T4YRMEM/graph.json","fetch_events":"https://pith.science/api/pith-number/BPEYYDR2PZT2KDHN5S2T4YRMEM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM/action/storage_attestation","attest_author":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM/action/author_attestation","sign_citation":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM/action/citation_signature","submit_replication":"https://pith.science/pith/BPEYYDR2PZT2KDHN5S2T4YRMEM/action/replication_record"}},"created_at":"2026-07-05T11:55:04.333278+00:00","updated_at":"2026-07-05T11:55:04.333278+00:00"}