{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5NAHHZAA4YRWZDKQT3HU2DBC7O","short_pith_number":"pith:5NAHHZAA","schema_version":"1.0","canonical_sha256":"eb4073e400e6236c8d509ecf4d0c22fb929afcd3e353f310a03f35d55d645365","source":{"kind":"arxiv","id":"2512.03042","version":3},"attestation_state":"computed","paper":{"title":"PPTArena: A Benchmark for PowerPoint Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Liang-Yan Gui, Michael Ofengenden, Yunze Man, Yu-Xiong Wang, Ziqi Pang","submitted_at":"2025-12-02T18:59:50Z","abstract_excerpt":"We introduce PPTArena, a benchmark for PowerPoint editing that evaluates how agents modify real slides from natural-language instructions. Unlike benchmarks that rely on image-PDF renderings or text-to-slide generation, PPTArena features 100 decks with over 1,300 human-curated edits across 2,125 slides, spanning text, charts, animations, and professional master styles. Each edit pairs a ground-truth deck with a target rubric and is scored by two Vision-Language Model (VLM) judges: one rates instruction following from structural diffs, the other visual quality from slide images. On top of this "},"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":"2512.03042","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-12-02T18:59:50Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"32e2ecd480b39f5e9b2b67fcc8a6ddc1ec35eabaffa1885d151f6369d3f6b3dc","abstract_canon_sha256":"a0e23e425ca31eafb4cb015943959192f3dd49b29deee57676ec3dabf6e047bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T00:16:51.467051Z","signature_b64":"8PJlTLT5bsMX+ZqG03mfLCtaxF42/hwP6TXIdBpGtdgI/G5kD07z+gDMzmhotuZtvJd5BopLDdUuHklsg/XHAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eb4073e400e6236c8d509ecf4d0c22fb929afcd3e353f310a03f35d55d645365","last_reissued_at":"2026-07-03T00:16:51.466538Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T00:16:51.466538Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PPTArena: A Benchmark for PowerPoint Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Liang-Yan Gui, Michael Ofengenden, Yunze Man, Yu-Xiong Wang, Ziqi Pang","submitted_at":"2025-12-02T18:59:50Z","abstract_excerpt":"We introduce PPTArena, a benchmark for PowerPoint editing that evaluates how agents modify real slides from natural-language instructions. Unlike benchmarks that rely on image-PDF renderings or text-to-slide generation, PPTArena features 100 decks with over 1,300 human-curated edits across 2,125 slides, spanning text, charts, animations, and professional master styles. Each edit pairs a ground-truth deck with a target rubric and is scored by two Vision-Language Model (VLM) judges: one rates instruction following from structural diffs, the other visual quality from slide images. On top of this "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.03042","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/2512.03042/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":"2512.03042","created_at":"2026-07-03T00:16:51.466593+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.03042v3","created_at":"2026-07-03T00:16:51.466593+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.03042","created_at":"2026-07-03T00:16:51.466593+00:00"},{"alias_kind":"pith_short_12","alias_value":"5NAHHZAA4YRW","created_at":"2026-07-03T00:16:51.466593+00:00"},{"alias_kind":"pith_short_16","alias_value":"5NAHHZAA4YRWZDKQ","created_at":"2026-07-03T00:16:51.466593+00:00"},{"alias_kind":"pith_short_8","alias_value":"5NAHHZAA","created_at":"2026-07-03T00:16:51.466593+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2606.19256","citing_title":"X+Slides: Benchmarking Audience-Conditioned Slide Generation","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.28775","citing_title":"Learn from Weaknesses: Automated Domain Specialization for Small Computer-Use Agents","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15202","citing_title":"DeepSlide: From Artifacts to Presentation Delivery","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15597","citing_title":"LLMs Corrupt Your Documents When You Delegate","ref_index":67,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O","json":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O.json","graph_json":"https://pith.science/api/pith-number/5NAHHZAA4YRWZDKQT3HU2DBC7O/graph.json","events_json":"https://pith.science/api/pith-number/5NAHHZAA4YRWZDKQT3HU2DBC7O/events.json","paper":"https://pith.science/paper/5NAHHZAA"},"agent_actions":{"view_html":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O","download_json":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O.json","view_paper":"https://pith.science/paper/5NAHHZAA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.03042&json=true","fetch_graph":"https://pith.science/api/pith-number/5NAHHZAA4YRWZDKQT3HU2DBC7O/graph.json","fetch_events":"https://pith.science/api/pith-number/5NAHHZAA4YRWZDKQT3HU2DBC7O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O/action/storage_attestation","attest_author":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O/action/author_attestation","sign_citation":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O/action/citation_signature","submit_replication":"https://pith.science/pith/5NAHHZAA4YRWZDKQT3HU2DBC7O/action/replication_record"}},"created_at":"2026-07-03T00:16:51.466593+00:00","updated_at":"2026-07-03T00:16:51.466593+00:00"}