{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5M2BSPTDL6TGUIMFS5XEI2GOUC","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":"cd0fc2afe74aa97ab83be67e8e57cc41d91bf83935d6afa9a39f8a1310f1be58","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T02:11:56Z","title_canon_sha256":"d82357d2f9c57c163f9e9fe87b16d7b730748ed4c1e73e210cac613170b2574e"},"schema_version":"1.0","source":{"id":"2505.21893","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21893","created_at":"2026-05-20T00:05:29Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21893v3","created_at":"2026-05-20T00:05:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21893","created_at":"2026-05-20T00:05:29Z"},{"alias_kind":"pith_short_12","alias_value":"5M2BSPTDL6TG","created_at":"2026-05-20T00:05:29Z"},{"alias_kind":"pith_short_16","alias_value":"5M2BSPTDL6TGUIMF","created_at":"2026-05-20T00:05:29Z"},{"alias_kind":"pith_short_8","alias_value":"5M2BSPTD","created_at":"2026-05-20T00:05:29Z"}],"graph_snapshots":[{"event_id":"sha256:e4be393b46f160ff8a7453614b68aed50392c6f47964d3099e991dc0516df21e","target":"graph","created_at":"2026-05-20T00:05:29Z","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/2505.21893/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Preference learning has garnered extensive attention as an effective technique for aligning diffusion models with human preferences in visual generation. However, existing alignment approaches such as Diffusion-DPO suffer from two fundamental challenges: training instability caused by high gradient variances at various timesteps and high parameter sensitivities, and off-policy bias arising from the discrepancy between the optimization data and the policy models' distribution. Our first contribution is a systematic analysis of diffusion trajectories across different timesteps, identifying that ","authors_text":"Hao Li, Junyan Wang, Mengping Yang, Xiaomeng Yang, Zhijian Zhou, Zhiyu Tan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T02:11:56Z","title":"SIPO: Stabilized and Improved Preference Optimization for Aligning Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21893","kind":"arxiv","version":3},"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":"sha256:31c5a4f86b31c9e267bd1461cc6b2ae193e39f8cd58b1e6e85dd2dcba7b548c2","target":"record","created_at":"2026-05-20T00:05:29Z","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":"cd0fc2afe74aa97ab83be67e8e57cc41d91bf83935d6afa9a39f8a1310f1be58","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-28T02:11:56Z","title_canon_sha256":"d82357d2f9c57c163f9e9fe87b16d7b730748ed4c1e73e210cac613170b2574e"},"schema_version":"1.0","source":{"id":"2505.21893","kind":"arxiv","version":3}},"canonical_sha256":"eb34193e635fa66a2185976e4468cea09392f0056f53c537a59cd2b71608497d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eb34193e635fa66a2185976e4468cea09392f0056f53c537a59cd2b71608497d","first_computed_at":"2026-05-20T00:05:29.310650Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:05:29.310650Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"s+OJWAgLqj/i4SxN129QBmLN5/oOu51b+DVVPzKwMRhbXtxbqJmDml/4Ji3aHVqNQLHNBMmo31hMUeeJezPIBA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:05:29.311276Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21893","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:31c5a4f86b31c9e267bd1461cc6b2ae193e39f8cd58b1e6e85dd2dcba7b548c2","sha256:e4be393b46f160ff8a7453614b68aed50392c6f47964d3099e991dc0516df21e"],"state_sha256":"1c2384be46ad7b2f037b97b69d13f91babb8c009967337424b399cd8d80ab1e0"}