{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HUNVO5KQQWMUFOYZBSABM2ACOA","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":"ed21ecbe4c16978bc4cc0c4b52444d042d443d967df93a45ac6d8bce5c85130f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T11:14:18Z","title_canon_sha256":"33bb592aa691709011a3424ff6fa021ffd3a630b83d0994051e6aae9fffe466d"},"schema_version":"1.0","source":{"id":"2605.16937","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16937","created_at":"2026-05-20T00:03:31Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16937v1","created_at":"2026-05-20T00:03:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16937","created_at":"2026-05-20T00:03:31Z"},{"alias_kind":"pith_short_12","alias_value":"HUNVO5KQQWMU","created_at":"2026-05-20T00:03:31Z"},{"alias_kind":"pith_short_16","alias_value":"HUNVO5KQQWMUFOYZ","created_at":"2026-05-20T00:03:31Z"},{"alias_kind":"pith_short_8","alias_value":"HUNVO5KQ","created_at":"2026-05-20T00:03:31Z"}],"graph_snapshots":[{"event_id":"sha256:05de6b8f7d4c8b6a07a4fd0fec9c1af0ea430fb4d4cda41d0ab502b6df3dae7e","target":"graph","created_at":"2026-05-20T00:03:31Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"we propose Dynamic Extreme VIew Synthesis-GRPO (DEVIS-GRPO), a GRPO-based framework for trajectory-controlled video generation, the first online policy gradient method for extreme view video generation. Central to our approach is a novel sampling strategy: Accumulative Dynamic Extreme VIew Synthesis (ADEVIS), which achieves large-view camera motions by progressively accumulating small-view increments."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"Progressively accumulating small-view increments reliably produces high-quality large-view motions and increased sampling diversity without introducing artifacts that the multi-level reward cannot filter, eliminating the need for expensive paired large-view videos."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"DEVIS-GRPO applies online policy gradients with an accumulative small-to-large view sampling strategy and multi-level rewards to improve trajectory-controlled extreme view video generation, reporting gains on Kubric-4D, iPhone, and DL3DV datasets."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Accumulating small camera increments during sampling lets a policy-gradient model handle extreme-view video generation without paired large-motion training data."}],"snapshot_sha256":"d3d57330badd617e3885d0355c4d0a09b483071150c08be4944325541650e26f"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"eadbb74667398b7eb4f92594ff7ebaa82c1cf2dfe6ca686753956c12f7b71bc3"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.133724Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:00:54.351634Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"cited_work_retraction","ran_at":"2026-05-19T20:21:57.751616Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T18:41:56.249228Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.331368Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16937/integrity.json","findings":[],"snapshot_sha256":"c8b9238b24030c0644a7a437b1689948db13854e7c3e03fc09e36ad9cb5b9843","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Trajectory-controlled video generation has become essential for controllable video generation. While current methods perform well under small-view camera motions, they degrade significantly with large-view motions. Existing solutions for extreme-view synthesis typically require dedicated video pairs, demanding substantial annotation effort. To address these limitations, we propose Dynamic Extreme VIew Synthesis-GRPO (DEVIS-GRPO), a GRPO-based framework for trajectory-controlled video generation, the first online policy gradient method for extreme view video generation. Central to our approach ","authors_text":"Fang Liu, Huimin Wu, Licheng Jiao, Lingling Li, Qing Li, Yi Zuo","cross_cats":[],"headline":"Accumulating small camera increments during sampling lets a policy-gradient model handle extreme-view video generation without paired large-motion training data.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T11:14:18Z","title":"DEVIS-GRPO: Unleashing GRPO on Dynamic Extreme View Synthesis"},"references":{"count":79,"internal_anchors":14,"resolved_work":79,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages =","work_id":"b4323590-3548-479f-9bdb-84b233131806","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Cami2v: Camera-controlled image-to-video diffu- sion model","work_id":"dfc5b040-8e44-47be-8714-65cc5b5f8400","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Lindell and Sergey Tulyakov , booktitle =","work_id":"3da69074-d191-45c2-ae9e-b478ef8800c7","year":2025},{"cited_arxiv_id":"2404.02101","doi":"","is_internal_anchor":true,"ref_index":4,"title":"CameraCtrl: Enabling Camera Control for Text-to-Video Generation","work_id":"1c05c278-c023-4ef0-a359-25a41f1065eb","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages =","work_id":"735560ec-d143-465b-885e-689a29b5dd34","year":null}],"snapshot_sha256":"db1f2730fd84bf5ebdf9e75a53fd8286440fccef68866355081898f7228af8ac"},"source":{"id":"2605.16937","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T20:50:48.173269Z","id":"3124d67b-a514-422e-a945-2d730636bdbc","model_set":{"reader":"grok-4.3"},"one_line_summary":"DEVIS-GRPO applies online policy gradients with an accumulative small-to-large view sampling strategy and multi-level rewards to improve trajectory-controlled extreme view video generation, reporting gains on Kubric-4D, iPhone, and DL3DV datasets.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Accumulating small camera increments during sampling lets a policy-gradient model handle extreme-view video generation without paired large-motion training data.","strongest_claim":"we propose Dynamic Extreme VIew Synthesis-GRPO (DEVIS-GRPO), a GRPO-based framework for trajectory-controlled video generation, the first online policy gradient method for extreme view video generation. Central to our approach is a novel sampling strategy: Accumulative Dynamic Extreme VIew Synthesis (ADEVIS), which achieves large-view camera motions by progressively accumulating small-view increments.","weakest_assumption":"Progressively accumulating small-view increments reliably produces high-quality large-view motions and increased sampling diversity without introducing artifacts that the multi-level reward cannot filter, eliminating the need for expensive paired large-view videos."}},"verdict_id":"3124d67b-a514-422e-a945-2d730636bdbc"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3982915805bb9573842508186944cbfa7157cff792160d9fe9175a5ee5fa88dd","target":"record","created_at":"2026-05-20T00:03:31Z","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":"ed21ecbe4c16978bc4cc0c4b52444d042d443d967df93a45ac6d8bce5c85130f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-16T11:14:18Z","title_canon_sha256":"33bb592aa691709011a3424ff6fa021ffd3a630b83d0994051e6aae9fffe466d"},"schema_version":"1.0","source":{"id":"2605.16937","kind":"arxiv","version":1}},"canonical_sha256":"3d1b577550859942bb190c801668027001d05739d5da11c4d5d3c2bfacf1b265","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d1b577550859942bb190c801668027001d05739d5da11c4d5d3c2bfacf1b265","first_computed_at":"2026-05-20T00:03:31.822922Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:31.822922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tVvLtZm6xTDS3PJBMICUQ10PIJ1dRm6ClcVvJVptHGnyW63OJ91mRvUfKMmjUaScqOmW6olS6tvOOZupg4EeAQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:31.823732Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16937","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3982915805bb9573842508186944cbfa7157cff792160d9fe9175a5ee5fa88dd","sha256:05de6b8f7d4c8b6a07a4fd0fec9c1af0ea430fb4d4cda41d0ab502b6df3dae7e"],"state_sha256":"48e0baaeedc3a75db8d01b44d09e65d8d69730aded11e64bdac0c5b8122d3150"}