{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:MH37KOEWQ3CKUMTTLXZ2VSJYAW","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":"42855c5e6a6eb95d3bac3ca76d010eb41d49d615788f648354d6e9fa9f393538","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-06T15:33:32Z","title_canon_sha256":"96d9ddfd618163e25087a85f7791848e06e16b619a6cd70eb89205ba29437570"},"schema_version":"1.0","source":{"id":"2403.03788","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.03788","created_at":"2026-07-05T07:53:01Z"},{"alias_kind":"arxiv_version","alias_value":"2403.03788v1","created_at":"2026-07-05T07:53:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03788","created_at":"2026-07-05T07:53:01Z"},{"alias_kind":"pith_short_12","alias_value":"MH37KOEWQ3CK","created_at":"2026-07-05T07:53:01Z"},{"alias_kind":"pith_short_16","alias_value":"MH37KOEWQ3CKUMTT","created_at":"2026-07-05T07:53:01Z"},{"alias_kind":"pith_short_8","alias_value":"MH37KOEW","created_at":"2026-07-05T07:53:01Z"}],"graph_snapshots":[{"event_id":"sha256:4d4417b484dfe9f5f06370ce50d6893ab0dbdfd295c9479e83dfb372dd21bc44","target":"graph","created_at":"2026-07-05T07:53:01Z","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/2403.03788/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The growing dependence on Large Language Models (LLMs) for finishing user instructions necessitates a comprehensive understanding of their robustness to complex task completion in real-world situations. To address this critical need, we propose the PowerPoint Task Completion Robustness benchmark (PPTC-R) to measure LLMs' robustness to the user PPT task instruction and software version. Specifically, we construct adversarial user instructions by attacking user instructions at sentence, semantic, and multi-language levels. To assess the robustness of Language Models to software versions, we vary","authors_text":"Dongyan Zhao, Nan Duan, Yaobo Liang, Yiduo Guo, Zekai Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-06T15:33:32Z","title":"PPTC-R benchmark: Towards Evaluating the Robustness of Large Language Models for PowerPoint Task Completion"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03788","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":"sha256:55e33b2c58be76b15a0679a7bb230f80785e9e32bb340b2fc74647c1b8793a7f","target":"record","created_at":"2026-07-05T07:53:01Z","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":"42855c5e6a6eb95d3bac3ca76d010eb41d49d615788f648354d6e9fa9f393538","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-06T15:33:32Z","title_canon_sha256":"96d9ddfd618163e25087a85f7791848e06e16b619a6cd70eb89205ba29437570"},"schema_version":"1.0","source":{"id":"2403.03788","kind":"arxiv","version":1}},"canonical_sha256":"61f7f5389686c4aa32735df3aac93805b95eacb6d6f9a598f289ea0d5618dfed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"61f7f5389686c4aa32735df3aac93805b95eacb6d6f9a598f289ea0d5618dfed","first_computed_at":"2026-07-05T07:53:01.571552Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:53:01.571552Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hejE2HZkx8583ff/nmLcyKrMyjC+n6b3LG4y+UCs6a+g4qN3pdT3bXlSvFzDe8JzWvKNBBJdbnSGtN6blfGLBA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:53:01.571903Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.03788","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:55e33b2c58be76b15a0679a7bb230f80785e9e32bb340b2fc74647c1b8793a7f","sha256:4d4417b484dfe9f5f06370ce50d6893ab0dbdfd295c9479e83dfb372dd21bc44"],"state_sha256":"4adac852827270ac365ea7addb1237b3eead869bcd16dc0035379f673e437a96"}