{"as_of":"2026-08-17T15:17:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:a7c7f9857b065c200c905396ff1a58b6c9f1711b82f837462edf6fc2ce54628b","coverage":[{"denominator":37,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":37,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-04T22:58:08.865854Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+00:00","state":"measured"},{"denominator":9,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":9,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-04T14:42:14.456588Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-06-30T18:14:59.829960Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":"2509.06942","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-06-30T18:14:59.829960Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":"94bda585-7a9f-4b25-8338-9bfbe86c83bf","year":2025},"citing_paper":{"arxiv_id":"2509.23951","last_updated":"2026-06-26T07:53:46Z","snapshot_observed_at":"2026-08-16T05:43:37.647093Z","submitted_at":"2025-09-28T16:14:10Z","title":"HunyuanImage 3.0 Technical Report","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-05-16T02:02:32.806844Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2509.23951"},"observation_digest":"sha256:91285a5a10ee95f3018244d8648dba27e718e8afd1839248561aa87c26f86fee","observation_id":"050f2db6-1155-40b3-addd-b4a38dca59bf","resolution":{"observed_at":"2026-05-16T02:02:32.875975Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-08-04T14:42:14.456588Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference.arXiv preprint arXiv:2509.06942, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.23951","last_updated":"2026-06-26T07:53:46Z","snapshot_observed_at":"2026-08-16T05:43:37.647093Z","submitted_at":"2025-09-28T16:14:10Z","title":"HunyuanImage 3.0 Technical Report","version":3},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-04T14:42:14.456588Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2509.23951"},"observation_digest":"sha256:45fd5e3379e38ce917bfb9667de286c19673e698b60a05eda065a3f0a01b9925","observation_id":"75ffebf9-39d7-455f-a228-277a91f4a04c","resolution":{"observed_at":"2026-08-04T14:42:14.456588Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-08-03T21:47:20.079741Z","title":"Directly aligning the full diffusion tra- jectory with fine-grained human preference.arXiv preprint arXiv:2509.06942, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2511.13649","last_updated":"2026-07-03T19:09:38Z","snapshot_observed_at":"2026-08-16T18:41:17.237259Z","submitted_at":"2025-11-17T17:59:54Z","title":"Distribution Matching Distillation Meets Reinforcement Learning","version":5},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-03T21:47:20.079741Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2511.13649"},"observation_digest":"sha256:da4251d6fabf29f06be6982bc1bcfb844f4d07dd3a18cf13e4fab72f5be4ab48","observation_id":"36adf26e-5a56-4acb-b6b1-11c20e671a27","resolution":{"observed_at":"2026-08-03T21:47:20.079741Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-08-03T11:36:14.589083Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.05729","last_updated":"2026-05-26T12:06:13Z","snapshot_observed_at":"2026-08-14T21:01:19.475876Z","submitted_at":"2026-01-09T11:15:27Z","title":"TAGRPO: Boosting GRPO on Image-to-Video Generation with Direct Trajectory Alignment","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-03T11:36:14.589083Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2601.05729"},"observation_digest":"sha256:604d99708069f01b886ed79149e2c15cd61de161067c3a92efa4ce762f96651e","observation_id":"5e2f6747-0baf-421d-ace6-06ded4e6f846","resolution":{"observed_at":"2026-08-03T11:36:14.589083Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-08-02T23:59:17.056636Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2602.12155","last_updated":"2026-06-29T06:08:55Z","snapshot_observed_at":"2026-08-14T16:33:50.814086Z","submitted_at":"2026-02-12T16:36:33Z","title":"FAIL: Flow Matching Adversarial Imitation Learning for Image Generation","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-02T23:59:17.056636Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2602.12155"},"observation_digest":"sha256:47ae1221b9ecdc15a06c8c3947296ff5c74ae07a51feef87866784f3fe44969c","observation_id":"3226be7c-d242-46f4-a027-d3641c51b797","resolution":{"observed_at":"2026-08-02T23:59:17.056636Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":"2509.06942","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-06-30T18:14:59.829960Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":"94bda585-7a9f-4b25-8338-9bfbe86c83bf","year":2025},"citing_paper":{"arxiv_id":"2604.15311","last_updated":"2026-05-03T11:22:04Z","snapshot_observed_at":"2026-08-14T05:02:02.000375Z","submitted_at":"2026-04-16T17:59:56Z","title":"LeapAlign: Post-Training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-05-10T11:23:28.424453Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2604.15311"},"observation_digest":"sha256:8db58f01736992273a6a348ce08afc39fd3b95f4cfe771cd955ad7a2b937e2c8","observation_id":"8cda3702-eab3-4ec1-be36-9373d8cc1d2d","resolution":{"observed_at":"2026-05-10T11:25:18.802132Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":"2509.06942","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-06-30T18:14:59.829960Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":"94bda585-7a9f-4b25-8338-9bfbe86c83bf","year":2025},"citing_paper":{"arxiv_id":"2605.19839","last_updated":"2026-06-04T12:21:01Z","snapshot_observed_at":"2026-07-06T23:30:30.141040Z","submitted_at":"2026-05-19T13:35:11Z","title":"When Preference Labels Fall Short: Aligning Diffusion Models from Real Data","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-05-20T06:13:01.821585Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2605.19839"},"observation_digest":"sha256:75b817032223426ca19f53e6a4da7d70f0ff2a0686c667d845c858c39efb5625","observation_id":"d5fa8438-4f51-41fb-a518-a42ffbd777e5","resolution":{"observed_at":"2026-05-20T06:13:05.225166Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":"2509.06942","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-06-30T18:14:59.829960Z","title":"Directly aligning the full diffusion trajectory with fine-grained human preference","venue":null,"work_id":"94bda585-7a9f-4b25-8338-9bfbe86c83bf","year":2025},"citing_paper":{"arxiv_id":"2605.19839","last_updated":"2026-06-04T12:21:01Z","snapshot_observed_at":"2026-07-06T23:30:30.141040Z","submitted_at":"2026-05-19T13:35:11Z","title":"When Preference Labels Fall Short: Aligning Diffusion Models from Real Data","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-06-30T18:12:11.972394Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2605.19839"},"observation_digest":"sha256:f14d8f459f81eab3f8a97dba1ec0ad84b596ca74707d72e6132cde30fbe7d6ef","observation_id":"469810ea-6acd-4b6d-b676-8de9a8b02b7b","resolution":{"observed_at":"2026-06-30T18:14:59.831627Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2509.06942","snapshot_observed_at":"2026-08-01T11:22:44.090146Z","title":"arXiv preprint arXiv:2509.06942 (2025) 8","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19923","last_updated":"2026-07-22T08:55:52Z","snapshot_observed_at":"2026-08-15T18:58:57.636160Z","submitted_at":"2026-07-22T08:55:52Z","title":"WearWow: Native 2K Multi-Garment Virtual Try-On via Adaptive Token Packing and Preference Alignment","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-01T11:22:44.090146Z"},"links":{"cited_paper":"/paper/2509.06942","citing_paper":"/paper/2607.19923"},"observation_digest":"sha256:14cd27005927b8f89a6aeadd5e4506d48a96131efac5296c92ccb6a66d040c14","observation_id":"81f85514-ed9c-4283-a6e2-1aa71b2e527e","resolution":{"observed_at":"2026-08-01T11:22:44.090146Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2509.06942/citation-record","integrity":"/paper/2509.06942/integrity","json":"/paper/2509.06942/citation-record.json","paper":"/paper/2509.06942"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.194823Z","title":"com / discus0434/aesthetic-predictor-v2-5, 2025","venue":null,"work_id":"4337c64f-5287-4c4d-8bab-ca5d821d9dd7","year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.753725Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:ab4d640fd3eb0c060c8b58ed7144f9c64a08d6f81e9222bb6198584cdcc3d583","observation_id":"52070015-3c8d-4ab5-a1af-f88d128c5838","resolution":{"observed_at":"2026-08-04T22:58:09.198025Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2303.08797","last_updated":"2025-10-09T00:43:44Z","snapshot_observed_at":"2026-07-06T15:03:52.079498Z","submitted_at":"2023-03-15T17:43:42Z","title":"Stochastic Interpolants: A Unifying Framework for Flows and Diffusions","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.08797","snapshot_observed_at":"2026-08-04T22:58:08.757460Z","title":"Stochastic interpolants: A unifying framework for flows and diffusions.arXiv preprint arXiv:2303.08797,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.757460Z"},"links":{"cited_paper":"/paper/2303.08797","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:f388393df49e97cd7c364a5437cb1254e17865071d8d1ae8c9de37d2fcfdedea","observation_id":"937e5ece-ff4a-48be-acc5-38fc07c01ef5","resolution":{"observed_at":"2026-08-04T22:58:08.757460Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2507.19002","last_updated":"2025-07-25T07:01:50Z","snapshot_observed_at":"2026-08-17T12:55:05.888354Z","submitted_at":"2025-07-25T07:01:50Z","title":"Enhancing Reward Models for High-quality Image Generation: Beyond Text-Image Alignment","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.19002","snapshot_observed_at":"2026-08-04T22:58:08.761058Z","title":"Enhancing reward models for high-quality image generation: Beyond text-image align- ment.arXiv preprint arXiv:2507.19002, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.761058Z"},"links":{"cited_paper":"/paper/2507.19002","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:567c3cd6b709934c7c6fad916b20c64f06650d4953e46ce00cc0d8088f8e8b3d","observation_id":"dde1f6ee-f58c-4ec4-829d-5167820bbb4f","resolution":{"observed_at":"2026-08-04T22:58:08.761058Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.13301","last_updated":"2024-01-04T19:11:25Z","snapshot_observed_at":"2026-08-09T16:20:23.732146Z","submitted_at":"2023-05-22T17:57:41Z","title":"Training Diffusion Models with Reinforcement Learning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.13301","snapshot_observed_at":"2026-08-04T22:58:08.764401Z","title":"Training diffusion models with reinforce- ment learning.arXiv preprint arXiv:2305.13301, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.764401Z"},"links":{"cited_paper":"/paper/2305.13301","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:861fbff83753934b87c0ae0d93519fb54dc565fa2d2b25996e5a601e1835b577","observation_id":"e1241b49-4de3-4534-9528-cc47770100df","resolution":{"observed_at":"2026-08-04T22:58:08.764401Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.17400","last_updated":"2024-06-21T16:45:11Z","snapshot_observed_at":"2026-08-06T10:08:54.816113Z","submitted_at":"2023-09-29T17:01:02Z","title":"Directly Fine-Tuning Diffusion Models on Differentiable Rewards","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.17400","snapshot_observed_at":"2026-08-04T22:58:08.767666Z","title":"Directly fine-tuning diffusion models on differentiable re- wards.arXiv preprint arXiv:2309.17400, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.767666Z"},"links":{"cited_paper":"/paper/2309.17400","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:2758c31cbb87bdc3849f19e16663c56233da46846609f4bc78568ae618e99a3f","observation_id":"dc7c3733-7051-454a-926e-41d618b2efd2","resolution":{"observed_at":"2026-08-04T22:58:08.767666Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.08861","last_updated":"2025-01-07T18:12:27Z","snapshot_observed_at":"2026-08-16T13:18:47.535778Z","submitted_at":"2024-09-13T14:22:14Z","title":"Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Control","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.08861","snapshot_observed_at":"2026-08-04T22:58:08.771147Z","title":"Adjoint matching: Fine-tuning flow and diffusion generative models with memoryless stochastic op- timal control.arXiv preprint arXiv:2409.08861, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.771147Z"},"links":{"cited_paper":"/paper/2409.08861","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:22c06a6b8c5bffa883b737b2c07e3ddfc1311df4c8f79e2c74136415babd55e1","observation_id":"2411b54e-52c0-4eb5-ac85-12e3d374c44f","resolution":{"observed_at":"2026-08-04T22:58:08.771147Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.13362","last_updated":"2024-09-19T20:42:34Z","snapshot_observed_at":"2026-08-16T15:58:59.222684Z","submitted_at":"2023-01-31T01:37:48Z","title":"Optimizing DDPM Sampling with Shortcut Fine-Tuning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.13362","snapshot_observed_at":"2026-08-04T22:58:08.774404Z","title":"Optimizing ddpm sampling with shortcut fine-tuning.arXiv preprint arXiv:2301.13362,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.774404Z"},"links":{"cited_paper":"/paper/2301.13362","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:f77c4913bcb0de55003b4880a9f79f8ede5ae26159a34b1f288a865748e70997","observation_id":"e48e195f-45f6-486e-89d3-81e54981447b","resolution":{"observed_at":"2026-08-04T22:58:08.774404Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:08.777600Z","title":"Dpok: Reinforcement learning for fine-tuning text-to-image diffu- sion models.Advances in Neural Information Processing Systems, 36:79858–79885, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.777600Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:c20d11cc95bb674f32df2b95112842d72ebf2629be9af7c7e07af6381454f035","observation_id":"f589bae5-fdda-4c81-a1c6-1ffd31060a87","resolution":{"observed_at":"2026-08-04T22:58:08.777600Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.180184Z","title":"Geneval: An object-focused framework for evaluating text- to-image alignment.Advances in Neural Information Pro- cessing Systems, 36:52132–52152, 2023","venue":null,"work_id":"d8982a17-f024-41d8-9509-75fbab0fe168","year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.780775Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:0238c3d52d471a6308814a1a95c48a567e0fffbf78f206afbe48f8e40fba06a5","observation_id":"bb0f0370-442b-4d34-a567-954c21fd7969","resolution":{"observed_at":"2026-08-04T22:58:09.183500Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2207.12598","last_updated":"2022-07-26T01:42:07Z","snapshot_observed_at":"2026-08-14T06:37:15.299690Z","submitted_at":"2022-07-26T01:42:07Z","title":"Classifier-Free Diffusion Guidance","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2207.12598","snapshot_observed_at":"2026-08-04T22:58:08.783694Z","title":"Classifier-free diffusion guidance.arXiv preprint arXiv:2207.12598, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.783694Z"},"links":{"cited_paper":"/paper/2207.12598","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:9c72834c203ecd6c9ac9f39fea3e735ff37deddd8266136c8c499e2158098252","observation_id":"4269ba9b-8c0a-4483-b8fd-70af9a9ea9e6","resolution":{"observed_at":"2026-08-04T22:58:08.783694Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:08.786888Z","title":"Denoising dif- fusion probabilistic models.Advances in neural information processing systems, 33:6840–6851, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.786888Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:c5835853396e6fc7a6d8d33f1b266125d166b912f59379830321711fc0442e2c","observation_id":"8b99b031-d8cb-4512-9ddd-c019e57a7870","resolution":{"observed_at":"2026-08-04T22:58:08.786888Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.164856Z","title":"Pick-a-pic: An open dataset of user preferences for text-to-image generation","venue":null,"work_id":"bfa20b68-a31b-41e5-8530-8571a07e6740","year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.789768Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:7b2af4b58544ed78381ca70c8fbaf01f56f5f1fd8697ca33c48d19942908d7c2","observation_id":"90e7f48e-887d-42a6-ab7b-44768cc3bb14","resolution":{"observed_at":"2026-08-04T22:58:09.168349Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.155368Z","title":"Flux.https://github.com/ black-forest-labs/flux, 2024","venue":null,"work_id":"72e1b867-5b30-47c1-958e-3fa4bae9a2b5","year":2024},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.792737Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:2bb55265e3e12bb0c3b596297f289bb043f871f311f3e7c9441a0b489141c3ab","observation_id":"7d0ff485-5249-493a-b495-b1a3cb9de8d6","resolution":{"observed_at":"2026-08-04T22:58:09.158617Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2302.12192","last_updated":"2023-02-23T17:34:53Z","snapshot_observed_at":"2026-07-06T14:55:12.100148Z","submitted_at":"2023-02-23T17:34:53Z","title":"Aligning Text-to-Image Models using Human Feedback","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.12192","snapshot_observed_at":"2026-08-04T22:58:08.795495Z","title":"Aligning text- to-image models using human feedback.arXiv preprint arXiv:2302.12192, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.795495Z"},"links":{"cited_paper":"/paper/2302.12192","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:1ef1051447c73cca12268d12706dbe95ed05408afbcef8ab5706f87e7dfadedc","observation_id":"4611180b-1710-4ecf-bcb3-445a55c2831f","resolution":{"observed_at":"2026-08-04T22:58:08.795495Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.146068Z","title":"Flux.1 krea [dev].https://github.com/krea-ai/flux- krea, 2025","venue":null,"work_id":"8a70c3f5-dac7-4d1d-96a1-1c2ecf2bf498","year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.798546Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:3667dcde1229860c06cf2f825368697b009b7cfee1f66e0635248409e046f057","observation_id":"91ccb4a6-a8a6-4951-b6df-9aa612c6462d","resolution":{"observed_at":"2026-08-04T22:58:09.149435Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.08748","last_updated":"2024-05-14T16:33:25Z","snapshot_observed_at":"2026-07-06T18:14:16.386835Z","submitted_at":"2024-05-14T16:33:25Z","title":"Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.08748","snapshot_observed_at":"2026-08-04T22:58:08.801305Z","title":"Hunyuan-dit: A powerful multi-resolution diffusion transformer with fine-grained chi- nese understanding.arXiv preprint arXiv:2405.08748, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.801305Z"},"links":{"cited_paper":"/paper/2405.08748","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:632a72f27b4c456347bdd287742c6db7dec9d97a59d1b00a187d83c7dff88f6e","observation_id":"2aea1fa6-9b79-49d5-af67-7cdfd5a622c4","resolution":{"observed_at":"2026-08-04T22:58:08.801305Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.136634Z","title":"Aes- thetic post-training diffusion models from generic prefer- ences with step-by-step preference optimization","venue":null,"work_id":"1634de17-633b-48c6-89af-a158072a1d85","year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.804355Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:553092f9231a1f86d3c67a26be3ec57c8bce52edc22c3434d41330e3c1cc8484","observation_id":"2a38496e-c1c6-4958-b816-bbf7733a7eca","resolution":{"observed_at":"2026-08-04T22:58:09.140049Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2210.02747","last_updated":"2023-02-08T15:46:05Z","snapshot_observed_at":"2026-08-16T02:30:42.660030Z","submitted_at":"2022-10-06T08:32:20Z","title":"Flow Matching for Generative Modeling","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.02747","snapshot_observed_at":"2026-08-04T22:58:08.807313Z","title":"Flow matching for generative mod- eling.arXiv preprint arXiv:2210.02747, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.807313Z"},"links":{"cited_paper":"/paper/2210.02747","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:5d1d0ad015dbe83dbb649aef3e68c4d982ef75f74e1850393fb364b6d73f5460","observation_id":"7591b143-3683-43f6-b2ab-59b3d5465bf7","resolution":{"observed_at":"2026-08-04T22:58:08.807313Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2505.05470","last_updated":"2025-10-27T09:57:02Z","snapshot_observed_at":"2026-08-13T10:15:00.293616Z","submitted_at":"2025-05-08T17:58:45Z","title":"Flow-GRPO: Training Flow Matching Models via Online RL","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.05470","snapshot_observed_at":"2026-08-04T22:58:08.810494Z","title":"Flow-grpo: Training flow matching models via on- line rl.arXiv preprint arXiv:2505.05470, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.810494Z"},"links":{"cited_paper":"/paper/2505.05470","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:bf984982d7ae4da03a415f2b466a7504bdf54154c959dca3da3b24893c528da3","observation_id":"9a6dc78c-e6a1-4063-a01f-2df826adbc25","resolution":{"observed_at":"2026-08-04T22:58:08.810494Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2209.03003","last_updated":"2022-09-07T08:59:55Z","snapshot_observed_at":"2026-07-06T13:49:40.974495Z","submitted_at":"2022-09-07T08:59:55Z","title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2209.03003","snapshot_observed_at":"2026-08-04T22:58:08.813669Z","title":"Flow straight and fast: Learning to generate and transfer data with rectified flow.arXiv preprint arXiv:2209.03003, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.813669Z"},"links":{"cited_paper":"/paper/2209.03003","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:a9a898c61d9ca6b785aa9a955b970e9e56ffa03f0f08e958c64af3c6cddef3da","observation_id":"5f6d553e-f857-401a-827c-1b250f7dd7cd","resolution":{"observed_at":"2026-08-04T22:58:08.813669Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.127107Z","title":"Sit: Explor- ing flow and diffusion-based generative models with scalable interpolant transformers","venue":null,"work_id":"68216f8e-dd60-4c5d-8301-2f814c7d7f01","year":2024},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.816730Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:a2d099276bbb954324d1b0d038e8554c0df98b2d178d37dec5ac9b67cdabf302","observation_id":"25000ad8-be8a-4dff-be0e-cc99ea05a220","resolution":{"observed_at":"2026-08-04T22:58:09.130470Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2508.03789","last_updated":"2025-08-22T08:53:37Z","snapshot_observed_at":"2026-08-16T17:02:25.152374Z","submitted_at":"2025-08-05T17:17:13Z","title":"HPSv3: Towards Wide-Spectrum Human Preference Score","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.03789","snapshot_observed_at":"2026-08-04T22:58:08.819616Z","title":"Hpsv3: Towards wide-spectrum human preference score","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.819616Z"},"links":{"cited_paper":"/paper/2508.03789","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:5128e198a3c17614f3fb3db12d6c030bf9004e117c873d1b434294ee0be14bae","observation_id":"5134029c-8262-4629-9416-3e3bcc9275c4","resolution":{"observed_at":"2026-08-04T22:58:08.819616Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.03544","last_updated":"2022-02-14T09:05:38Z","snapshot_observed_at":"2026-08-16T03:45:29.671274Z","submitted_at":"2022-01-10T18:58:52Z","title":"The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.03544","snapshot_observed_at":"2026-08-04T22:58:08.822693Z","title":"The ef- fects of reward misspecification: Mapping and mitigating misaligned models.arXiv preprint arXiv:2201.03544, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.822693Z"},"links":{"cited_paper":"/paper/2201.03544","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:beeba1c38fc125475ec2cfacb663e3fa93ff7933a39b9a6d6598a72b56ffbb58","observation_id":"9f810d7d-4446-4519-a6b0-2f31a88963f6","resolution":{"observed_at":"2026-08-04T22:58:08.822693Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.117418Z","title":"Aligning text-to-image diffusion models with reward backpropagation","venue":null,"work_id":"5ab985f1-9584-458e-b3c3-4bc0a1fb2554","year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.825751Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:0c2dba827b61fccfa4dd9064082511d36c90bc40b9414c6951daf13d2bf5c9d0","observation_id":"b0a10205-d2cf-419a-ae24-971bee4740a6","resolution":{"observed_at":"2026-08-04T22:58:09.120814Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.08737","last_updated":"2024-07-11T17:59:45Z","snapshot_observed_at":"2026-08-16T13:34:58.945033Z","submitted_at":"2024-07-11T17:59:45Z","title":"Video Diffusion Alignment via Reward Gradients","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.08737","snapshot_observed_at":"2026-08-04T22:58:08.828632Z","title":"Video diffusion align- ment via reward gradients.arXiv preprint arXiv:2407.08737,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.828632Z"},"links":{"cited_paper":"/paper/2407.08737","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:1caf98a2acd7d7c0520a0cd01af7d9dbe97d7c65711c9d3b6b0d9b02618fa91f","observation_id":"4193fa4e-2e0f-43f8-b61f-d6714ea31c85","resolution":{"observed_at":"2026-08-04T22:58:08.828632Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:08.831915Z","title":"Learning transferable visual models from natural language supervi- sion","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.831915Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:131220c24b84bcb2163d604b5b0ca639521229c351e68738aba06405cb12c12a","observation_id":"180c8dc2-073e-4b28-bd09-c7620c33f9f5","resolution":{"observed_at":"2026-08-04T22:58:08.831915Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.102907Z","title":"Laion-aesthetics.https : / / laion.ai/blog/laion- aesthetics/, 2022","venue":null,"work_id":"ab679087-a8a6-4522-bbf0-f1900f75eaf6","year":2022},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.834994Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:987a6c52d808726400d7585b5b4757f8432e1fd0d6ec0ce177637ac0ddd639d8","observation_id":"2b34f3f7-57b5-4fac-a0dc-f429582478a7","resolution":{"observed_at":"2026-08-04T22:58:09.106260Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2010.02502","last_updated":"2022-10-05T20:19:21Z","snapshot_observed_at":"2026-08-11T15:38:14.931716Z","submitted_at":"2020-10-06T06:15:51Z","title":"Denoising Diffusion Implicit Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.02502","snapshot_observed_at":"2026-08-04T22:58:08.837795Z","title":"Denoising diffusion implicit models.arXiv preprint arXiv:2010.02502, 2020","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.837795Z"},"links":{"cited_paper":"/paper/2010.02502","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:ca9f80f45b50b5ab07b5efd248ca5232e83cdb4f75bdea4f64b33291c47246ab","observation_id":"f702561d-4610-46cb-b88f-fe00408b790c","resolution":{"observed_at":"2026-08-04T22:58:08.837795Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2011.13456","last_updated":"2021-02-10T18:17:04Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-11-26T19:39:10Z","title":"Score-Based Generative Modeling through Stochastic Differential Equations","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2011.13456","snapshot_observed_at":"2026-08-04T22:58:08.841022Z","title":"Score-based generative modeling through stochastic differential equa- tions.arXiv preprint arXiv:2011.13456, 2020","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.841022Z"},"links":{"cited_paper":"/paper/2011.13456","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:d838d4906055ee36bceac7ee8fb99e7894030631e85467278e25348874920541","observation_id":"fffbbe72-75d6-4031-a142-e7a511c6efdb","resolution":{"observed_at":"2026-08-04T22:58:08.841022Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.20751","last_updated":"2026-04-20T06:12:09Z","snapshot_observed_at":"2026-08-15T08:35:23.989525Z","submitted_at":"2025-08-28T13:11:24Z","title":"Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.20751","snapshot_observed_at":"2026-08-04T22:58:08.844072Z","title":"Pref-grpo: Pairwise preference reward-based grpo for sta- ble text-to-image reinforcement learning.arXiv preprint arXiv:2508.20751, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.844072Z"},"links":{"cited_paper":"/paper/2508.20751","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:c4ef243dc85f798a0f021866ee8884d0673cb0b89989aca5adcb47eccf6709ee","observation_id":"009cc442-5440-4708-b446-54d7a68e7e11","resolution":{"observed_at":"2026-08-04T22:58:08.844072Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.14896","last_updated":"2023-07-06T11:53:19Z","snapshot_observed_at":"2026-08-16T16:20:53.427900Z","submitted_at":"2022-10-26T17:54:20Z","title":"DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.14896","snapshot_observed_at":"2026-08-04T22:58:08.847327Z","title":"Diffu- siondb: A large-scale prompt gallery dataset for text-to- image generative models.arXiv preprint arXiv:2210.14896,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.847327Z"},"links":{"cited_paper":"/paper/2210.14896","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:486f44aba6923ed8d50d390462eee9e8ae4c1b2954e97576893f00656389c2d5","observation_id":"c1cba328-40a4-45ef-9b35-ef8d610369e5","resolution":{"observed_at":"2026-08-04T22:58:08.847327Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.093449Z","title":"Human preference score: Better aligning text- to-image models with human preference","venue":null,"work_id":"20523251-05d1-4697-ba6b-a89182d14c04","year":2096},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.850662Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:e44ea4cc2aa8b404c1331410413725684c8c1304f5a43e2ee852ccb806851608","observation_id":"5afdee4f-8697-4eb0-bbb2-bfe977d7516e","resolution":{"observed_at":"2026-08-04T22:58:09.096797Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.083386Z","title":"Imagere- ward: Learning and evaluating human preferences for text- to-image generation.Advances in Neural Information Pro- cessing Systems, 36:15903–15935, 2023","venue":null,"work_id":"ee4dd67c-c538-4bc1-ab5d-c4a392baa467","year":2023},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.853783Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:e2f901e9b4c7f9dee9fd866b9612e8847beaf8e09c6e4e9e661e99410ac455a0","observation_id":"10982d8d-9e14-4341-bbfc-9049c883b985","resolution":{"observed_at":"2026-08-04T22:58:09.086823Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2505.07818","last_updated":"2025-08-28T17:19:45Z","snapshot_observed_at":"2026-08-15T21:26:30.952106Z","submitted_at":"2025-05-12T17:59:34Z","title":"DanceGRPO: Unleashing GRPO on Visual Generation","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.07818","snapshot_observed_at":"2026-08-04T22:58:08.856771Z","title":"Dancegrpo: Unleashing grpo on visual generation.arXiv preprint arXiv:2505.07818, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.856771Z"},"links":{"cited_paper":"/paper/2505.07818","citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:56e3a9806a0c2213c397a3c1b6d6b0b01200dd287d64139d0a7e6fe6402452b3","observation_id":"b31f2086-0538-4936-8c03-06d9442abf05","resolution":{"observed_at":"2026-08-04T22:58:08.856771Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:08.859993Z","title":"Teaching large language models to regress accurate image quality scores using score distribution.arXiv preprint arXiv:2501.11561, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.859993Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:72acdcd2b6b788948dbeb7abf845f1340f04142daf4c4739c33ab93a340e5e78","observation_id":"c5448a8f-781c-4075-b2f7-9a8baaa2d00c","resolution":{"observed_at":"2026-08-04T22:58:08.859993Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:09.071727Z","title":"Learning multi- dimensional human preference for text-to-image generation","venue":null,"work_id":"5316bffd-6b7e-4b5c-81f1-ec3496b250dd","year":2024},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.862860Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:95ea31e003a01967f37030e9ac10a24e53ea8fb0b40d53e7446603a6b95efc00","observation_id":"d539de12-5a72-47e2-b3b4-db2babfb6f4a","resolution":{"observed_at":"2026-08-04T22:58:09.076792Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-04T22:58:08.865854Z","title":"Diffusion model as a noise-aware latent reward model for step-level preference optimization.arXiv preprint arXiv:2502.01051, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference","version":3},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-04T22:58:08.865854Z"},"links":{"citing_paper":"/paper/2509.06942"},"observation_digest":"sha256:99bc952258be7bd97b2472b7f260cc43f67f85f863de75abb5a2129fbb0e252e","observation_id":"2599237f-f1be-4309-9df1-f3a845038047","resolution":{"observed_at":"2026-08-04T22:58:08.865854Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2509.06942","last_updated":"2025-09-11T17:14:11Z","latest_version":3,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-14T05:10:53.900739Z","submitted_at":"2025-09-08T17:54:08Z","title":"Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference"},"reference_resolution":{"displayed":37,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":25,"verified_exact":0,"verified_fuzzy":12},"total_outbound_references":37},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 37 of 37 outbound references and 9 inbound Pith citation observations for arXiv:2509.06942."}