{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XUWX5X7GXPBLAH3FSOVIJ6G3Y3","short_pith_number":"pith:XUWX5X7G","canonical_record":{"source":{"id":"2505.23331","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T10:45:38Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0fe400851dc24b79af35e9267be809258730eef2d6ae21479a88b088e17ec2a4","abstract_canon_sha256":"8f0c0263b6004e8fa39d62c071556af0c51747f24f5883fd392d6e7cbb6213a8"},"schema_version":"1.0"},"canonical_sha256":"bd2d7edfe6bbc2b01f6593aa84f8dbc6ecfdd945753bec4fb73c1054f3a45184","source":{"kind":"arxiv","id":"2505.23331","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23331","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23331v2","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23331","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_12","alias_value":"XUWX5X7GXPBL","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_16","alias_value":"XUWX5X7GXPBLAH3F","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_8","alias_value":"XUWX5X7G","created_at":"2026-07-05T11:28:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XUWX5X7GXPBLAH3FSOVIJ6G3Y3","target":"record","payload":{"canonical_record":{"source":{"id":"2505.23331","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T10:45:38Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"0fe400851dc24b79af35e9267be809258730eef2d6ae21479a88b088e17ec2a4","abstract_canon_sha256":"8f0c0263b6004e8fa39d62c071556af0c51747f24f5883fd392d6e7cbb6213a8"},"schema_version":"1.0"},"canonical_sha256":"bd2d7edfe6bbc2b01f6593aa84f8dbc6ecfdd945753bec4fb73c1054f3a45184","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:28:36.758170Z","signature_b64":"T16cQ5bJmBH8F2eq/6eA62H0LxVVD8dAdGhHqTg+t6VgSriFns6IVzU/vAB648K+im2cJk+bV+5XBxfcz81KDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd2d7edfe6bbc2b01f6593aa84f8dbc6ecfdd945753bec4fb73c1054f3a45184","last_reissued_at":"2026-07-05T11:28:36.757662Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:28:36.757662Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.23331","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:28:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sBI2LZ/uMFqnM974v3546rcdJPgUo+zdGshVr+w9tw6ZgcR4v6Ue/AsHVQbPqRk/c6cHTkCKWD7XjtDWvf+wAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T19:04:13.102231Z"},"content_sha256":"0eae98bf342e02073a889ed2e6b64a9130ac65c02089684b20351d74c3b0b153","schema_version":"1.0","event_id":"sha256:0eae98bf342e02073a889ed2e6b64a9130ac65c02089684b20351d74c3b0b153"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XUWX5X7GXPBLAH3FSOVIJ6G3Y3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fine-Tuning Next-Scale Visual Autoregressive Models with Group Relative Policy Optimization","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haitz S\\'aez de Oc\\'ariz Borde, Matteo Gallici","submitted_at":"2025-05-29T10:45:38Z","abstract_excerpt":"Fine-tuning pre-trained generative models with Reinforcement Learning (RL) has emerged as an effective approach for aligning outputs more closely with nuanced human preferences. In this paper, we investigate the application of Group Relative Policy Optimization (GRPO) to fine-tune next-scale visual autoregressive (VAR) models. Our empirical results demonstrate that this approach enables alignment to intricate reward signals derived from aesthetic predictors and CLIP embeddings, significantly enhancing image quality and enabling precise control over the generation style. Interestingly, by lever"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23331","kind":"arxiv","version":2},"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/2505.23331/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":"2026-07-05T11:28:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4JPk+qGEU4pZTOiNpvR85Km4OqqUHAXbrkQ8NGTRIL/rnfCJgKotiEHT/2cKMw7jpp/KoviieT58OL/hxuoAAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T19:04:13.102751Z"},"content_sha256":"5fe660af26088fc1314e9af7f6a0f5850165c9a426e960568075dff5816a51ce","schema_version":"1.0","event_id":"sha256:5fe660af26088fc1314e9af7f6a0f5850165c9a426e960568075dff5816a51ce"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/bundle.json","state_url":"https://pith.science/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/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":"2026-08-19T19:04:13Z","links":{"resolver":"https://pith.science/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3","bundle":"https://pith.science/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/bundle.json","state":"https://pith.science/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XUWX5X7GXPBLAH3FSOVIJ6G3Y3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XUWX5X7GXPBLAH3FSOVIJ6G3Y3","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":"8f0c0263b6004e8fa39d62c071556af0c51747f24f5883fd392d6e7cbb6213a8","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T10:45:38Z","title_canon_sha256":"0fe400851dc24b79af35e9267be809258730eef2d6ae21479a88b088e17ec2a4"},"schema_version":"1.0","source":{"id":"2505.23331","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23331","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23331v2","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23331","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_12","alias_value":"XUWX5X7GXPBL","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_16","alias_value":"XUWX5X7GXPBLAH3F","created_at":"2026-07-05T11:28:36Z"},{"alias_kind":"pith_short_8","alias_value":"XUWX5X7G","created_at":"2026-07-05T11:28:36Z"}],"graph_snapshots":[{"event_id":"sha256:5fe660af26088fc1314e9af7f6a0f5850165c9a426e960568075dff5816a51ce","target":"graph","created_at":"2026-07-05T11:28:36Z","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/2505.23331/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning pre-trained generative models with Reinforcement Learning (RL) has emerged as an effective approach for aligning outputs more closely with nuanced human preferences. In this paper, we investigate the application of Group Relative Policy Optimization (GRPO) to fine-tune next-scale visual autoregressive (VAR) models. Our empirical results demonstrate that this approach enables alignment to intricate reward signals derived from aesthetic predictors and CLIP embeddings, significantly enhancing image quality and enabling precise control over the generation style. Interestingly, by lever","authors_text":"Haitz S\\'aez de Oc\\'ariz Borde, Matteo Gallici","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T10:45:38Z","title":"Fine-Tuning Next-Scale Visual Autoregressive Models with Group Relative Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23331","kind":"arxiv","version":2},"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":"sha256:0eae98bf342e02073a889ed2e6b64a9130ac65c02089684b20351d74c3b0b153","target":"record","created_at":"2026-07-05T11:28:36Z","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":"8f0c0263b6004e8fa39d62c071556af0c51747f24f5883fd392d6e7cbb6213a8","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T10:45:38Z","title_canon_sha256":"0fe400851dc24b79af35e9267be809258730eef2d6ae21479a88b088e17ec2a4"},"schema_version":"1.0","source":{"id":"2505.23331","kind":"arxiv","version":2}},"canonical_sha256":"bd2d7edfe6bbc2b01f6593aa84f8dbc6ecfdd945753bec4fb73c1054f3a45184","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bd2d7edfe6bbc2b01f6593aa84f8dbc6ecfdd945753bec4fb73c1054f3a45184","first_computed_at":"2026-07-05T11:28:36.757662Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:28:36.757662Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"T16cQ5bJmBH8F2eq/6eA62H0LxVVD8dAdGhHqTg+t6VgSriFns6IVzU/vAB648K+im2cJk+bV+5XBxfcz81KDw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:28:36.758170Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.23331","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0eae98bf342e02073a889ed2e6b64a9130ac65c02089684b20351d74c3b0b153","sha256:5fe660af26088fc1314e9af7f6a0f5850165c9a426e960568075dff5816a51ce"],"state_sha256":"fbb7fc76fe989ed9f5528954d70c4a10dd93cd586cfa411224c1ada09cbb4b39"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H3uSLKqqVUtZKV3OXhINZoSQytwvFZuGBS0qXwGVz8ol79bnepM/jOYlEfQuPG1rHLqCfJ02XGtromroUW5lDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T19:04:13.108907Z","bundle_sha256":"13d77e9ccf7507e5efd033f83da3738ed17560d75c9d2a511fd3722763cc74a6"}}