{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:7MOJFQSGUZS5INDMRIMD43S4PZ","short_pith_number":"pith:7MOJFQSG","canonical_record":{"source":{"id":"2508.06924","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-09T10:37:26Z","cross_cats_sorted":[],"title_canon_sha256":"3b94ae56d4d125c07275acb006ed073e930d437d8098530ccef0361860b56d08","abstract_canon_sha256":"395d1d64461ca27651f3eb39d2397ae281d19e741d0bb7822834e5a3d1c1caf8"},"schema_version":"1.0"},"canonical_sha256":"fb1c92c246a665d4346c8a183e6e5c7e7f0a8d0998d49fe359c42a8e658258af","source":{"kind":"arxiv","id":"2508.06924","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06924","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06924v1","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06924","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_12","alias_value":"7MOJFQSGUZS5","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_16","alias_value":"7MOJFQSGUZS5INDM","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_8","alias_value":"7MOJFQSG","created_at":"2026-07-05T11:51:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:7MOJFQSGUZS5INDMRIMD43S4PZ","target":"record","payload":{"canonical_record":{"source":{"id":"2508.06924","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-09T10:37:26Z","cross_cats_sorted":[],"title_canon_sha256":"3b94ae56d4d125c07275acb006ed073e930d437d8098530ccef0361860b56d08","abstract_canon_sha256":"395d1d64461ca27651f3eb39d2397ae281d19e741d0bb7822834e5a3d1c1caf8"},"schema_version":"1.0"},"canonical_sha256":"fb1c92c246a665d4346c8a183e6e5c7e7f0a8d0998d49fe359c42a8e658258af","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:37.702222Z","signature_b64":"ZiUdJZBC9lwnGTibQBCRblJY1Re86l5RS/LECyGA8jrZ2eH+1dfDUKL9Zk6BQT08GHin27/Z52Nt1WrTMoIIAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb1c92c246a665d4346c8a183e6e5c7e7f0a8d0998d49fe359c42a8e658258af","last_reissued_at":"2026-07-05T11:51:37.701734Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:37.701734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.06924","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":"2026-07-05T11:51:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D38mdgq+B/xt9T/xy9gkY3mbsNZQ4FHC35sg3BQQTKwEn5CvKtvA/K/JNNWmF1ACc9aqBMBggFTpIexdEeerDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T11:43:37.207551Z"},"content_sha256":"4630db1f387e507dc8b6950705a1746b7d5f454cfb6515f9c826f78cb19b92dc","schema_version":"1.0","event_id":"sha256:4630db1f387e507dc8b6950705a1746b7d5f454cfb6515f9c826f78cb19b92dc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:7MOJFQSGUZS5INDMRIMD43S4PZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fuzheng Zhang, Guorui Zhou, Jingyuan Zhang, Qi Wang, Shihao Yuan, Wangmeng Zuo, Yahui Liu, Yang Yue","submitted_at":"2025-08-09T10:37:26Z","abstract_excerpt":"Inspired by the success of reinforcement learning (RL) in refining large language models (LLMs), we propose AR-GRPO, an approach to integrate online RL training into autoregressive (AR) image generation models. We adapt the Group Relative Policy Optimization (GRPO) algorithm to refine the vanilla autoregressive models' outputs by carefully designed reward functions that evaluate generated images across multiple quality dimensions, including perceptual quality, realism, and semantic fidelity. We conduct comprehensive experiments on both class-conditional (i.e., class-to-image) and text-conditio"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06924","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.06924/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":"2026-07-05T11:51:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5jbKbzaM19Ih5AJFPT9GpzZpOEytT47vo6PF3FasbpuJpxk+F3yoiYAhwP3Xz2ecfX+rrhm1LGfGsN0W+r9EBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T11:43:37.208083Z"},"content_sha256":"00449c8e4a6880ac311fa64191e411ab22363809faa15e2b5c0b9d81c3142845","schema_version":"1.0","event_id":"sha256:00449c8e4a6880ac311fa64191e411ab22363809faa15e2b5c0b9d81c3142845"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/bundle.json","state_url":"https://pith.science/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/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":"2026-08-05T11:43:37Z","links":{"resolver":"https://pith.science/pith/7MOJFQSGUZS5INDMRIMD43S4PZ","bundle":"https://pith.science/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/bundle.json","state":"https://pith.science/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7MOJFQSGUZS5INDMRIMD43S4PZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7MOJFQSGUZS5INDMRIMD43S4PZ","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":"395d1d64461ca27651f3eb39d2397ae281d19e741d0bb7822834e5a3d1c1caf8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-09T10:37:26Z","title_canon_sha256":"3b94ae56d4d125c07275acb006ed073e930d437d8098530ccef0361860b56d08"},"schema_version":"1.0","source":{"id":"2508.06924","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06924","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06924v1","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06924","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_12","alias_value":"7MOJFQSGUZS5","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_16","alias_value":"7MOJFQSGUZS5INDM","created_at":"2026-07-05T11:51:37Z"},{"alias_kind":"pith_short_8","alias_value":"7MOJFQSG","created_at":"2026-07-05T11:51:37Z"}],"graph_snapshots":[{"event_id":"sha256:00449c8e4a6880ac311fa64191e411ab22363809faa15e2b5c0b9d81c3142845","target":"graph","created_at":"2026-07-05T11:51:37Z","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/2508.06924/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Inspired by the success of reinforcement learning (RL) in refining large language models (LLMs), we propose AR-GRPO, an approach to integrate online RL training into autoregressive (AR) image generation models. We adapt the Group Relative Policy Optimization (GRPO) algorithm to refine the vanilla autoregressive models' outputs by carefully designed reward functions that evaluate generated images across multiple quality dimensions, including perceptual quality, realism, and semantic fidelity. We conduct comprehensive experiments on both class-conditional (i.e., class-to-image) and text-conditio","authors_text":"Fuzheng Zhang, Guorui Zhou, Jingyuan Zhang, Qi Wang, Shihao Yuan, Wangmeng Zuo, Yahui Liu, Yang Yue","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-09T10:37:26Z","title":"AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06924","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":"sha256:4630db1f387e507dc8b6950705a1746b7d5f454cfb6515f9c826f78cb19b92dc","target":"record","created_at":"2026-07-05T11:51:37Z","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":"395d1d64461ca27651f3eb39d2397ae281d19e741d0bb7822834e5a3d1c1caf8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-09T10:37:26Z","title_canon_sha256":"3b94ae56d4d125c07275acb006ed073e930d437d8098530ccef0361860b56d08"},"schema_version":"1.0","source":{"id":"2508.06924","kind":"arxiv","version":1}},"canonical_sha256":"fb1c92c246a665d4346c8a183e6e5c7e7f0a8d0998d49fe359c42a8e658258af","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fb1c92c246a665d4346c8a183e6e5c7e7f0a8d0998d49fe359c42a8e658258af","first_computed_at":"2026-07-05T11:51:37.701734Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:51:37.701734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZiUdJZBC9lwnGTibQBCRblJY1Re86l5RS/LECyGA8jrZ2eH+1dfDUKL9Zk6BQT08GHin27/Z52Nt1WrTMoIIAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:51:37.702222Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.06924","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4630db1f387e507dc8b6950705a1746b7d5f454cfb6515f9c826f78cb19b92dc","sha256:00449c8e4a6880ac311fa64191e411ab22363809faa15e2b5c0b9d81c3142845"],"state_sha256":"3d5e5a27cb40c649ef9b99d2580c36ee4eee9bcec82469f4426b06f84953e447"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yZhLMJoXyvqL0T6+aTmARMFdxdjXXA0PbNBE6N4XpPNludxvyQCkcCcGK49rxj/YHD/KITxCzzYdBAF3JJGVAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T11:43:37.211976Z","bundle_sha256":"ad5c93d5b5ef11ee4b8e3437c33e21e375195abffe1f4b106800712242694dd7"}}