{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:SQKJEUHT7NUGIUIGX7CO6HVTOG","short_pith_number":"pith:SQKJEUHT","canonical_record":{"source":{"id":"2503.18013","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-23T10:21:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fde22600d754d7905e49e34ef3e6293f30d7ac60127819cafad6e7e4b05a85ca","abstract_canon_sha256":"d99ceb87c4d3374785b5626180647d8b8ce17771f7fe9b3ad18d7273cbc853c7"},"schema_version":"1.0"},"canonical_sha256":"94149250f3fb68645106bfc4ef1eb3719b0a7efbe9b3bc2df7b5cc1acad4ebe5","source":{"kind":"arxiv","id":"2503.18013","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.18013","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"arxiv_version","alias_value":"2503.18013v1","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18013","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_12","alias_value":"SQKJEUHT7NUG","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_16","alias_value":"SQKJEUHT7NUGIUIG","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_8","alias_value":"SQKJEUHT","created_at":"2026-07-05T10:38:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:SQKJEUHT7NUGIUIGX7CO6HVTOG","target":"record","payload":{"canonical_record":{"source":{"id":"2503.18013","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-23T10:21:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fde22600d754d7905e49e34ef3e6293f30d7ac60127819cafad6e7e4b05a85ca","abstract_canon_sha256":"d99ceb87c4d3374785b5626180647d8b8ce17771f7fe9b3ad18d7273cbc853c7"},"schema_version":"1.0"},"canonical_sha256":"94149250f3fb68645106bfc4ef1eb3719b0a7efbe9b3bc2df7b5cc1acad4ebe5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:00.442871Z","signature_b64":"fxbbn7K0GNIUh05gZ7luwxcfKvntDvmZyAa2esOsgdJkheZ+eCEka6jEu2nXBOVjKjEMtZvO73nS/ITyA2jkBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94149250f3fb68645106bfc4ef1eb3719b0a7efbe9b3bc2df7b5cc1acad4ebe5","last_reissued_at":"2026-07-05T10:38:00.442372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:00.442372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.18013","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":"2026-07-05T10:38:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"59kVxWEacQS1Yb0+NBIULKf81mtJKuqjsrwS4wquxpqhbTqM0fc2wka67HwrSLydBIoPcXirXYjt0utQ17GBBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T13:20:18.093855Z"},"content_sha256":"7379cccc8766d7978a2b7b4df23940764d685fc5c2bb0a0be4428b345bb271a6","schema_version":"1.0","event_id":"sha256:7379cccc8766d7978a2b7b4df23940764d685fc5c2bb0a0be4428b345bb271a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:SQKJEUHT7NUGIUIGX7CO6HVTOG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Vision-R1: Evolving Human-Free Alignment in Large Vision-Language Models via Vision-Guided Reinforcement Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Fan Yang, Hongyin Zhao, Jinqiao Wang, Ming Tang, Shurong Zheng, Yousong Zhu, Yufei Zhan","submitted_at":"2025-03-23T10:21:14Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) typically follow a two-stage training paradigm-pretraining and supervised fine-tuning. Recently, preference optimization, derived from the language domain, has emerged as an effective post-training reinforcement strategy to enhance capabilities of LVLMs. However, constructing high-quality human-annotated preference data and developing robust reward models to mimic these preferences are both costly and challenging. Motivated by this observation, we propose Vision-R1, a novel vision-guided R1-like reinforcement learning algorithm for LVLMs that rewards models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18013","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/2503.18013/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":"2026-07-05T10:38:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lLMzt+2Mo7njkWcJKh1B5GZ7zydxYGWWBM0b/Z1gYK1aVk0Wxp7dKkIwAuYcAbHBlXXoEn5V2JXMGxcYao+iAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T13:20:18.094398Z"},"content_sha256":"bb42a063d1a7fde9c289967f9a8580c36716979a4d045855ae7685e0a904b019","schema_version":"1.0","event_id":"sha256:bb42a063d1a7fde9c289967f9a8580c36716979a4d045855ae7685e0a904b019"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/bundle.json","state_url":"https://pith.science/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/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":"2026-08-07T13:20:18Z","links":{"resolver":"https://pith.science/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG","bundle":"https://pith.science/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/bundle.json","state":"https://pith.science/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SQKJEUHT7NUGIUIGX7CO6HVTOG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SQKJEUHT7NUGIUIGX7CO6HVTOG","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":"d99ceb87c4d3374785b5626180647d8b8ce17771f7fe9b3ad18d7273cbc853c7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-23T10:21:14Z","title_canon_sha256":"fde22600d754d7905e49e34ef3e6293f30d7ac60127819cafad6e7e4b05a85ca"},"schema_version":"1.0","source":{"id":"2503.18013","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.18013","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"arxiv_version","alias_value":"2503.18013v1","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18013","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_12","alias_value":"SQKJEUHT7NUG","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_16","alias_value":"SQKJEUHT7NUGIUIG","created_at":"2026-07-05T10:38:00Z"},{"alias_kind":"pith_short_8","alias_value":"SQKJEUHT","created_at":"2026-07-05T10:38:00Z"}],"graph_snapshots":[{"event_id":"sha256:bb42a063d1a7fde9c289967f9a8580c36716979a4d045855ae7685e0a904b019","target":"graph","created_at":"2026-07-05T10:38:00Z","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/2503.18013/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Vision-Language Models (LVLMs) typically follow a two-stage training paradigm-pretraining and supervised fine-tuning. Recently, preference optimization, derived from the language domain, has emerged as an effective post-training reinforcement strategy to enhance capabilities of LVLMs. However, constructing high-quality human-annotated preference data and developing robust reward models to mimic these preferences are both costly and challenging. Motivated by this observation, we propose Vision-R1, a novel vision-guided R1-like reinforcement learning algorithm for LVLMs that rewards models","authors_text":"Fan Yang, Hongyin Zhao, Jinqiao Wang, Ming Tang, Shurong Zheng, Yousong Zhu, Yufei Zhan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-23T10:21:14Z","title":"Vision-R1: Evolving Human-Free Alignment in Large Vision-Language Models via Vision-Guided Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18013","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":"sha256:7379cccc8766d7978a2b7b4df23940764d685fc5c2bb0a0be4428b345bb271a6","target":"record","created_at":"2026-07-05T10:38:00Z","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":"d99ceb87c4d3374785b5626180647d8b8ce17771f7fe9b3ad18d7273cbc853c7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-23T10:21:14Z","title_canon_sha256":"fde22600d754d7905e49e34ef3e6293f30d7ac60127819cafad6e7e4b05a85ca"},"schema_version":"1.0","source":{"id":"2503.18013","kind":"arxiv","version":1}},"canonical_sha256":"94149250f3fb68645106bfc4ef1eb3719b0a7efbe9b3bc2df7b5cc1acad4ebe5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"94149250f3fb68645106bfc4ef1eb3719b0a7efbe9b3bc2df7b5cc1acad4ebe5","first_computed_at":"2026-07-05T10:38:00.442372Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:38:00.442372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fxbbn7K0GNIUh05gZ7luwxcfKvntDvmZyAa2esOsgdJkheZ+eCEka6jEu2nXBOVjKjEMtZvO73nS/ITyA2jkBg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:38:00.442871Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.18013","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7379cccc8766d7978a2b7b4df23940764d685fc5c2bb0a0be4428b345bb271a6","sha256:bb42a063d1a7fde9c289967f9a8580c36716979a4d045855ae7685e0a904b019"],"state_sha256":"cf5f7b9bfa09fd839a298d8c85d40ef2e29c656181a1825ba416e59f44bbd749"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cL+3zRAWG1Us2XWrLOj2Pwsg+uqdq4A8w3ydyUEN+3hACE+O8YhSrj1UFNRiTpkcBu0iIam/p6BhypsU6rcuCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T13:20:18.102302Z","bundle_sha256":"d4f76a5672bfea463f05a02279d9835aad9d8c98aa2c6188ac2850c36b51e964"}}