{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:I5PHEQH6KA4EHDDQ6R2Q6ZSO6X","short_pith_number":"pith:I5PHEQH6","canonical_record":{"source":{"id":"2608.06125","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-06T14:55:42Z","cross_cats_sorted":[],"title_canon_sha256":"13f75fa6739db102ca825b77adc78a4b037acc65e2bc7d7ad92ce8d961ce45f3","abstract_canon_sha256":"04f69c0876d6263716c9758bce7513749ac54b933c792067c07b9385af7b5ce7"},"schema_version":"1.0"},"canonical_sha256":"475e7240fe5038438c70f4750f664ef5d686d91ecd84b2344afb552d5ffe92d0","source":{"kind":"arxiv","id":"2608.06125","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.06125","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"arxiv_version","alias_value":"2608.06125v1","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06125","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_12","alias_value":"I5PHEQH6KA4E","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_16","alias_value":"I5PHEQH6KA4EHDDQ","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_8","alias_value":"I5PHEQH6","created_at":"2026-08-07T01:40:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:I5PHEQH6KA4EHDDQ6R2Q6ZSO6X","target":"record","payload":{"canonical_record":{"source":{"id":"2608.06125","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-06T14:55:42Z","cross_cats_sorted":[],"title_canon_sha256":"13f75fa6739db102ca825b77adc78a4b037acc65e2bc7d7ad92ce8d961ce45f3","abstract_canon_sha256":"04f69c0876d6263716c9758bce7513749ac54b933c792067c07b9385af7b5ce7"},"schema_version":"1.0"},"canonical_sha256":"475e7240fe5038438c70f4750f664ef5d686d91ecd84b2344afb552d5ffe92d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T01:40:35.768121Z","signature_b64":"aGWh5uVQpr+D6uScLx+cGsvvcZJpZkMzY61sj5kJk6gd0LpdScUzkHgaQHZjTGDk+p1NbzN6k44OcdM9ign1AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"475e7240fe5038438c70f4750f664ef5d686d91ecd84b2344afb552d5ffe92d0","last_reissued_at":"2026-08-07T01:40:35.766601Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T01:40:35.766601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.06125","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-07T01:40:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tapLWoqlGPQMrfA/JnIPyzXS7V0uEsxcFMgC94G9jhijOZUjomNE3iSzqdnAv/ZW57tY8V+tu4XBPP6lPfc2Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T17:26:11.495830Z"},"content_sha256":"5388c03cb4e6f49dddefc6960f2b2f69aeab49cc97aacaff56bddfe965a7f8da","schema_version":"1.0","event_id":"sha256:5388c03cb4e6f49dddefc6960f2b2f69aeab49cc97aacaff56bddfe965a7f8da"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:I5PHEQH6KA4EHDDQ6R2Q6ZSO6X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sample-Adaptive Latent Rewards for Uncertainty-Guided Diffusion Post-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chi Zhang, Haibin Huang, Ke Hao, Rui Li, Xuelong Li, Yuanzhi Liang, Ziqiao Weng","submitted_at":"2026-08-06T14:55:42Z","abstract_excerpt":"Latent reward models can supervise visual diffusion models without decoding intermediate states into pixel space. This makes alignment with human preferences more efficient. However, existing latent reward models output only scalar scores. They do not estimate the uncertainty of each prediction. The generator therefore cannot determine which feedback is reliable. This can drive optimization in the wrong direction and lead to reward hacking. We propose \\textsc{SURE}, a unified latent-space framework for image and video diffusion models. It learns reward distributions and directly uses their rel"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06125","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/2608.06125/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-07T01:40:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q1JtZlwTniRM3Pz28WWNcsK5wk443kcGruU42p5bYPq9zleBiYEfnsBi2YFJqGzwlDEjqU9jjkz/EApGQx5LDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T17:26:11.496394Z"},"content_sha256":"93da13063b7a9f1b0bf7da8ba4bb957a0fbc3daed8bb8f4ab8cdcdc1cdc7c2fe","schema_version":"1.0","event_id":"sha256:93da13063b7a9f1b0bf7da8ba4bb957a0fbc3daed8bb8f4ab8cdcdc1cdc7c2fe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/bundle.json","state_url":"https://pith.science/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-14T17:26:11Z","links":{"resolver":"https://pith.science/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X","bundle":"https://pith.science/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/bundle.json","state":"https://pith.science/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/I5PHEQH6KA4EHDDQ6R2Q6ZSO6X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:I5PHEQH6KA4EHDDQ6R2Q6ZSO6X","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"04f69c0876d6263716c9758bce7513749ac54b933c792067c07b9385af7b5ce7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-06T14:55:42Z","title_canon_sha256":"13f75fa6739db102ca825b77adc78a4b037acc65e2bc7d7ad92ce8d961ce45f3"},"schema_version":"1.0","source":{"id":"2608.06125","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.06125","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"arxiv_version","alias_value":"2608.06125v1","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06125","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_12","alias_value":"I5PHEQH6KA4E","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_16","alias_value":"I5PHEQH6KA4EHDDQ","created_at":"2026-08-07T01:40:35Z"},{"alias_kind":"pith_short_8","alias_value":"I5PHEQH6","created_at":"2026-08-07T01:40:35Z"}],"graph_snapshots":[{"event_id":"sha256:93da13063b7a9f1b0bf7da8ba4bb957a0fbc3daed8bb8f4ab8cdcdc1cdc7c2fe","target":"graph","created_at":"2026-08-07T01:40:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2608.06125/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Latent reward models can supervise visual diffusion models without decoding intermediate states into pixel space. This makes alignment with human preferences more efficient. However, existing latent reward models output only scalar scores. They do not estimate the uncertainty of each prediction. The generator therefore cannot determine which feedback is reliable. This can drive optimization in the wrong direction and lead to reward hacking. We propose \\textsc{SURE}, a unified latent-space framework for image and video diffusion models. It learns reward distributions and directly uses their rel","authors_text":"Chi Zhang, Haibin Huang, Ke Hao, Rui Li, Xuelong Li, Yuanzhi Liang, Ziqiao Weng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-06T14:55:42Z","title":"Sample-Adaptive Latent Rewards for Uncertainty-Guided Diffusion Post-Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06125","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5388c03cb4e6f49dddefc6960f2b2f69aeab49cc97aacaff56bddfe965a7f8da","target":"record","created_at":"2026-08-07T01:40:35Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"04f69c0876d6263716c9758bce7513749ac54b933c792067c07b9385af7b5ce7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-06T14:55:42Z","title_canon_sha256":"13f75fa6739db102ca825b77adc78a4b037acc65e2bc7d7ad92ce8d961ce45f3"},"schema_version":"1.0","source":{"id":"2608.06125","kind":"arxiv","version":1}},"canonical_sha256":"475e7240fe5038438c70f4750f664ef5d686d91ecd84b2344afb552d5ffe92d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"475e7240fe5038438c70f4750f664ef5d686d91ecd84b2344afb552d5ffe92d0","first_computed_at":"2026-08-07T01:40:35.766601Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-07T01:40:35.766601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aGWh5uVQpr+D6uScLx+cGsvvcZJpZkMzY61sj5kJk6gd0LpdScUzkHgaQHZjTGDk+p1NbzN6k44OcdM9ign1AQ==","signature_status":"signed_v1","signed_at":"2026-08-07T01:40:35.768121Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.06125","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5388c03cb4e6f49dddefc6960f2b2f69aeab49cc97aacaff56bddfe965a7f8da","sha256:93da13063b7a9f1b0bf7da8ba4bb957a0fbc3daed8bb8f4ab8cdcdc1cdc7c2fe"],"state_sha256":"c29efe89bda18ced801e1b61bd56209edbebf80258fca3a8d146af53b2383a9a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aFqgw7k9L4fhP3dnNcCRh4snHRh+IjiVapQC6HDMO68L9Vyf3lXdabZ1SpYocjZK8HZlsOfrZJrXC6lOcKt4Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T17:26:11.506752Z","bundle_sha256":"8b552af1def74801ed4c89d4661cb96af1678982c84e4da8c789db7f5d3a3fef"}}