{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KFV5LQB4UY2DWYMYR5E2O2H36E","short_pith_number":"pith:KFV5LQB4","schema_version":"1.0","canonical_sha256":"516bd5c03ca6343b61988f49a768fbf101d8b2e1725b764516a502a61c0375a3","source":{"kind":"arxiv","id":"2510.16888","version":3},"attestation_state":"computed","paper":{"title":"Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Feize Wu, Li Yuan, Qihui Zhang, Shaodong Wang, Shenghai Yuan, Wangbo Yu, Xinhua Cheng, Yang Ye, Yuwei Niu, Zheyuan Liu, Zhiyuan Yan, Zongjian Li","submitted_at":"2025-10-19T15:38:06Z","abstract_excerpt":"Instruction-based image editing has achieved remarkable progress; however, models solely trained via supervised fine-tuning often overfit to annotated patterns, hindering their ability to explore and generalize beyond training distributions. To this end, we introduce Edit-R1, a novel post-training framework for instruction-based image editing based on policy optimization. Specifically, we utilize Diffusion Negative-aware Finetuning (DiffusionNFT), a likelihood-free policy optimization method consistent with the flow matching forward process, thereby enabling the use of higher-order samplers an"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2510.16888","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-10-19T15:38:06Z","cross_cats_sorted":[],"title_canon_sha256":"ea1acdc528600813f1d4a566c58fb8d08c4f4c4fa16a1c82b87d79c59f78fdd3","abstract_canon_sha256":"3bd44088d8f3573ec5dc2ef4c549c981b80f9ad53eb46e53fabd2996a347b5d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T17:53:50.775832Z","signature_b64":"YspFofLpVq/hI6hDiB28icUuEfblWpVDHiE9AYGPFdKbMrY+KTmo1ML46dg8RS/5NIE/MQctmTSRCGqXHxXFBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"516bd5c03ca6343b61988f49a768fbf101d8b2e1725b764516a502a61c0375a3","last_reissued_at":"2026-05-21T17:53:50.773831Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T17:53:50.773831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Feize Wu, Li Yuan, Qihui Zhang, Shaodong Wang, Shenghai Yuan, Wangbo Yu, Xinhua Cheng, Yang Ye, Yuwei Niu, Zheyuan Liu, Zhiyuan Yan, Zongjian Li","submitted_at":"2025-10-19T15:38:06Z","abstract_excerpt":"Instruction-based image editing has achieved remarkable progress; however, models solely trained via supervised fine-tuning often overfit to annotated patterns, hindering their ability to explore and generalize beyond training distributions. To this end, we introduce Edit-R1, a novel post-training framework for instruction-based image editing based on policy optimization. Specifically, we utilize Diffusion Negative-aware Finetuning (DiffusionNFT), a likelihood-free policy optimization method consistent with the flow matching forward process, thereby enabling the use of higher-order samplers an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.16888","kind":"arxiv","version":3},"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/2510.16888/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2510.16888","created_at":"2026-05-21T17:53:50.773910+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.16888v3","created_at":"2026-05-21T17:53:50.773910+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.16888","created_at":"2026-05-21T17:53:50.773910+00:00"},{"alias_kind":"pith_short_12","alias_value":"KFV5LQB4UY2D","created_at":"2026-05-21T17:53:50.773910+00:00"},{"alias_kind":"pith_short_16","alias_value":"KFV5LQB4UY2DWYMY","created_at":"2026-05-21T17:53:50.773910+00:00"},{"alias_kind":"pith_short_8","alias_value":"KFV5LQB4","created_at":"2026-05-21T17:53:50.773910+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":23,"sample":[{"citing_arxiv_id":"2605.21605","citing_title":"GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2512.12598","citing_title":"Setting the Stage: Text-Driven Scene-Consistent Image Generation","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2602.23622","citing_title":"DLEBench: Evaluating Small-scale Object Editing Ability for Instruction-based Image Editing Model","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04128","citing_title":"JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18714","citing_title":"Semantic Generative Tuning for Unified Multimodal Models","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16951","citing_title":"Edit-GRPO: A Locality-Preserving Policy Optimization Framework for Image Editing","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2512.12675","citing_title":"Scone: Bridging Composition and Distinction in Subject-Driven Image Generation via Unified Understanding-Generation Modeling","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2503.07265","citing_title":"WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12724","citing_title":"Inline Critic Steers Image Editing","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.13062","citing_title":"Edit-Compass & EditReward-Compass: A Unified Benchmark for Image Editing and Reward Modeling","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12500","citing_title":"SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture","ref_index":72,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08354","citing_title":"Auto-Rubric as Reward: From Implicit Preferences to Explicit Multimodal Generative Criteria","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08703","citing_title":"RewardHarness: Self-Evolving Agentic Post-Training","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17565","citing_title":"UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10937","citing_title":"Power Reinforcement Post-Training of Text-to-Image Models with Super-Linear Advantage Shaping","ref_index":118,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04494","citing_title":"Towards General Preference Alignment: Diffusion Models at Nash Equilibrium","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2511.22699","citing_title":"Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19587","citing_title":"SmartPhotoCrafter: Unified Reasoning, Generation and Optimization for Automatic Photographic Image Editing","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2604.04911","citing_title":"SpatialEdit: Benchmarking Fine-Grained Image Spatial Editing","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15871","citing_title":"UniEditBench: A Unified and Cost-Effective Benchmark for Image and Video Editing via Distilled MLLMs","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17565","citing_title":"UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19009","citing_title":"Guiding Distribution Matching Distillation with Gradient-Based Reinforcement Learning","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04128","citing_title":"JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation","ref_index":46,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E","json":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E.json","graph_json":"https://pith.science/api/pith-number/KFV5LQB4UY2DWYMYR5E2O2H36E/graph.json","events_json":"https://pith.science/api/pith-number/KFV5LQB4UY2DWYMYR5E2O2H36E/events.json","paper":"https://pith.science/paper/KFV5LQB4"},"agent_actions":{"view_html":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E","download_json":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E.json","view_paper":"https://pith.science/paper/KFV5LQB4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.16888&json=true","fetch_graph":"https://pith.science/api/pith-number/KFV5LQB4UY2DWYMYR5E2O2H36E/graph.json","fetch_events":"https://pith.science/api/pith-number/KFV5LQB4UY2DWYMYR5E2O2H36E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E/action/storage_attestation","attest_author":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E/action/author_attestation","sign_citation":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E/action/citation_signature","submit_replication":"https://pith.science/pith/KFV5LQB4UY2DWYMYR5E2O2H36E/action/replication_record"}},"created_at":"2026-05-21T17:53:50.773910+00:00","updated_at":"2026-05-21T17:53:50.773910+00:00"}