{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:J7S7F6B6B6V4LY7KQQFJCUCDVI","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":"347f0534e8d5270a500a290d92a419ee0dc0aabdd4519a589dd557f9363a8779","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-16T17:55:54Z","title_canon_sha256":"d01a83c82d3baab872781e6961f50e0fb2d99bce51b39929d8452db2e7cb58df"},"schema_version":"1.0","source":{"id":"2505.11493","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11493","created_at":"2026-07-05T11:43:11Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11493v3","created_at":"2026-07-05T11:43:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11493","created_at":"2026-07-05T11:43:11Z"},{"alias_kind":"pith_short_12","alias_value":"J7S7F6B6B6V4","created_at":"2026-07-05T11:43:11Z"},{"alias_kind":"pith_short_16","alias_value":"J7S7F6B6B6V4LY7K","created_at":"2026-07-05T11:43:11Z"},{"alias_kind":"pith_short_8","alias_value":"J7S7F6B6","created_at":"2026-07-05T11:43:11Z"}],"graph_snapshots":[{"event_id":"sha256:3a4bc071182c84f757a63f142affb0c65f37329c79cf01afb4ae36d0575f0b30","target":"graph","created_at":"2026-07-05T11:43:11Z","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/2505.11493/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Editing images using natural language instructions has become a natural and expressive way to modify visual content; yet, evaluating the performance of such models remains challenging. Existing evaluation approaches often rely on image-text similarity metrics like CLIP, which lack precision. In this work, we introduce a new benchmark designed to evaluate text-guided image editing models in a more grounded manner, along two critical dimensions: (i) functional correctness, assessed via automatically generated multiple-choice questions that verify whether the intended change was successfully appl","authors_text":"Chen Chen, Jiasen Lu, Tsu-Jui Fu, Wenze Hu, Xinze Wang, Yinfei Yang, Yusu Qian, Zhe Gan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-16T17:55:54Z","title":"GIE-Bench: Towards Grounded Evaluation for Text-Guided Image Editing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11493","kind":"arxiv","version":3},"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":"sha256:10c0f5bedf0063185dba4088a2d63f7b81b26b0aa8a860369e276f74d47d3b4b","target":"record","created_at":"2026-07-05T11:43:11Z","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":"347f0534e8d5270a500a290d92a419ee0dc0aabdd4519a589dd557f9363a8779","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-16T17:55:54Z","title_canon_sha256":"d01a83c82d3baab872781e6961f50e0fb2d99bce51b39929d8452db2e7cb58df"},"schema_version":"1.0","source":{"id":"2505.11493","kind":"arxiv","version":3}},"canonical_sha256":"4fe5f2f83e0fabc5e3ea840a915043aa016dde961d65a3e269841e7e7e46d1c8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4fe5f2f83e0fabc5e3ea840a915043aa016dde961d65a3e269841e7e7e46d1c8","first_computed_at":"2026-07-05T11:43:11.917298Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:43:11.917298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lIYiI2TATbiq9m3DJJNmVX6LXmpuhjlf/wx4TAtMIwMgO5iDAunJWXfHEo0BWEoOAA8RVcspkaAocN1/lp6FAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:43:11.917804Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11493","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:10c0f5bedf0063185dba4088a2d63f7b81b26b0aa8a860369e276f74d47d3b4b","sha256:3a4bc071182c84f757a63f142affb0c65f37329c79cf01afb4ae36d0575f0b30"],"state_sha256":"fa81c03bcc2a79be321d76b5c18a3161aef5aa6e78b69cbbd9efcbab74995705"}