{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GRFS4NT7EJXXT3UOTO4OZTHNCO","short_pith_number":"pith:GRFS4NT7","schema_version":"1.0","canonical_sha256":"344b2e367f226f79ee8e9bb8eccced13bf3e0ec7bb56fad6dd8524f0954e20f0","source":{"kind":"arxiv","id":"2606.28301","version":1},"attestation_state":"computed","paper":{"title":"VGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DS","cs.NA","math.NA","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kijung Jeon, Molei Tao, Thuy-Duong Vuong","submitted_at":"2026-06-26T17:47:09Z","abstract_excerpt":"Inference-time scaling is a promising paradigm to improve generative models, especially when outputs must satisfy structural constraints or optimize downstream rewards. We consider Masked Diffusion Model (MDM) and introduce MDM-VGB, a discrete diffusion sampler that augments unmasking generation with theoretically principled reward-guided remasking. Inspired by the recent success of the classical Jerrum-Sinclair backtracking Markov chain in reward-tilted generation, MDM-VGB extends the backtracking random walk from a fixed prefix tree to a masked-state graph, allowing tokens to be unmasked and"},"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":"2606.28301","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-26T17:47:09Z","cross_cats_sorted":["cs.DS","cs.NA","math.NA","math.PR","stat.ML"],"title_canon_sha256":"cf4b36767f2e2ae0031ce849098663d037544ece402df46be8f68a46af59c698","abstract_canon_sha256":"7d80ccd711157d01f19d26323dce9718ba288965bc7920bae402da272f212809"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:15:12.475634Z","signature_b64":"v6kEwItz2fWUZ2TF9r0EUq2fb6/6iUrip8E242edoQihBkW8kY+hVMYIT0b2B3kJC4fTIG7AQdDn0cpzXKZuCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"344b2e367f226f79ee8e9bb8eccced13bf3e0ec7bb56fad6dd8524f0954e20f0","last_reissued_at":"2026-06-29T01:15:12.475248Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:15:12.475248Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DS","cs.NA","math.NA","math.PR","stat.ML"],"primary_cat":"cs.LG","authors_text":"Kijung Jeon, Molei Tao, Thuy-Duong Vuong","submitted_at":"2026-06-26T17:47:09Z","abstract_excerpt":"Inference-time scaling is a promising paradigm to improve generative models, especially when outputs must satisfy structural constraints or optimize downstream rewards. We consider Masked Diffusion Model (MDM) and introduce MDM-VGB, a discrete diffusion sampler that augments unmasking generation with theoretically principled reward-guided remasking. Inspired by the recent success of the classical Jerrum-Sinclair backtracking Markov chain in reward-tilted generation, MDM-VGB extends the backtracking random walk from a fixed prefix tree to a masked-state graph, allowing tokens to be unmasked and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28301","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/2606.28301/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":"2606.28301","created_at":"2026-06-29T01:15:12.475302+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.28301v1","created_at":"2026-06-29T01:15:12.475302+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28301","created_at":"2026-06-29T01:15:12.475302+00:00"},{"alias_kind":"pith_short_12","alias_value":"GRFS4NT7EJXX","created_at":"2026-06-29T01:15:12.475302+00:00"},{"alias_kind":"pith_short_16","alias_value":"GRFS4NT7EJXXT3UO","created_at":"2026-06-29T01:15:12.475302+00:00"},{"alias_kind":"pith_short_8","alias_value":"GRFS4NT7","created_at":"2026-06-29T01:15:12.475302+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO","json":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO.json","graph_json":"https://pith.science/api/pith-number/GRFS4NT7EJXXT3UOTO4OZTHNCO/graph.json","events_json":"https://pith.science/api/pith-number/GRFS4NT7EJXXT3UOTO4OZTHNCO/events.json","paper":"https://pith.science/paper/GRFS4NT7"},"agent_actions":{"view_html":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO","download_json":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO.json","view_paper":"https://pith.science/paper/GRFS4NT7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.28301&json=true","fetch_graph":"https://pith.science/api/pith-number/GRFS4NT7EJXXT3UOTO4OZTHNCO/graph.json","fetch_events":"https://pith.science/api/pith-number/GRFS4NT7EJXXT3UOTO4OZTHNCO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO/action/storage_attestation","attest_author":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO/action/author_attestation","sign_citation":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO/action/citation_signature","submit_replication":"https://pith.science/pith/GRFS4NT7EJXXT3UOTO4OZTHNCO/action/replication_record"}},"created_at":"2026-06-29T01:15:12.475302+00:00","updated_at":"2026-06-29T01:15:12.475302+00:00"}