{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:JLIPHWG3PMFRH6SXM6NIYK2CY2","short_pith_number":"pith:JLIPHWG3","canonical_record":{"source":{"id":"2505.24878","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T17:59:55Z","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"title_canon_sha256":"6d2ebec2c6f47a3b49213231b4aa5e6cf70c3cfebea8e12c46d01b9eccf668f9","abstract_canon_sha256":"e200f5d2b475883252c7d8483a7b2670548b01ed5907a2816b8b9bef12b0a702"},"schema_version":"1.0"},"canonical_sha256":"4ad0f3d8db7b0b13fa57679a8c2b42c693492f4a131e5984afff4a5606bac8df","source":{"kind":"arxiv","id":"2505.24878","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24878","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24878v1","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24878","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_12","alias_value":"JLIPHWG3PMFR","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_16","alias_value":"JLIPHWG3PMFRH6SX","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_8","alias_value":"JLIPHWG3","created_at":"2026-07-05T11:12:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:JLIPHWG3PMFRH6SXM6NIYK2CY2","target":"record","payload":{"canonical_record":{"source":{"id":"2505.24878","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T17:59:55Z","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"title_canon_sha256":"6d2ebec2c6f47a3b49213231b4aa5e6cf70c3cfebea8e12c46d01b9eccf668f9","abstract_canon_sha256":"e200f5d2b475883252c7d8483a7b2670548b01ed5907a2816b8b9bef12b0a702"},"schema_version":"1.0"},"canonical_sha256":"4ad0f3d8db7b0b13fa57679a8c2b42c693492f4a131e5984afff4a5606bac8df","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:58.229423Z","signature_b64":"76EAjVZMuXtFJA14hLj8hMWAsvluU1jAnc66+nDZ20qMH3xzI/INY0Om/2Zshqr45J8xfPRyztO7wSRf6yaEDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ad0f3d8db7b0b13fa57679a8c2b42c693492f4a131e5984afff4a5606bac8df","last_reissued_at":"2026-07-05T11:12:58.228869Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:58.228869Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.24878","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iRs9AlLuowWT8OfAsAmXKRmd+usvq+jKwm701zfkVWtYycjOmEQ2kyKgMkuLdik42jzGe1uxT2VpOdmAw6ThCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T21:29:46.924218Z"},"content_sha256":"7075293f43a26ab5e91e13d20e8475232638616f5a65d24b10582fb79e2f4228","schema_version":"1.0","event_id":"sha256:7075293f43a26ab5e91e13d20e8475232638616f5a65d24b10582fb79e2f4228"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:JLIPHWG3PMFRH6SXM6NIYK2CY2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Open CaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.CV","cs.LG"],"primary_cat":"cs.AI","authors_text":"Jiacheng Cui, Jiacheng Liu, Xiaohan Zhao, Yaxin Luo, Zhaoyi Li, Zhiqiang Shen","submitted_at":"2025-05-30T17:59:55Z","abstract_excerpt":"CAPTCHAs have been a critical bottleneck for deploying web agents in real-world applications, often blocking them from completing end-to-end automation tasks. While modern multimodal LLM agents have demonstrated impressive performance in static perception tasks, their ability to handle interactive, multi-step reasoning challenges like CAPTCHAs is largely untested. To address this gap, we introduce Open CaptchaWorld, the first web-based benchmark and platform specifically designed to evaluate the visual reasoning and interaction capabilities of MLLM-powered agents through diverse and dynamic CA"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24878","kind":"arxiv","version":1},"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/2505.24878/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:12:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZnN4ueR8XDYP2/gavNBCxoRhqjzotGguo19sAp8a0pDKaTJ1MYOCw1d7hqMAulyDwW3D7GnAXUOen3TV0XzvAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T21:29:46.924727Z"},"content_sha256":"7455fc5abd3bb6af4a6457f405c246bfdc92f70e53ef957190a0577a16a525a6","schema_version":"1.0","event_id":"sha256:7455fc5abd3bb6af4a6457f405c246bfdc92f70e53ef957190a0577a16a525a6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/bundle.json","state_url":"https://pith.science/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T21:29:46Z","links":{"resolver":"https://pith.science/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2","bundle":"https://pith.science/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/bundle.json","state":"https://pith.science/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JLIPHWG3PMFRH6SXM6NIYK2CY2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:JLIPHWG3PMFRH6SXM6NIYK2CY2","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":"e200f5d2b475883252c7d8483a7b2670548b01ed5907a2816b8b9bef12b0a702","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T17:59:55Z","title_canon_sha256":"6d2ebec2c6f47a3b49213231b4aa5e6cf70c3cfebea8e12c46d01b9eccf668f9"},"schema_version":"1.0","source":{"id":"2505.24878","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.24878","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"arxiv_version","alias_value":"2505.24878v1","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24878","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_12","alias_value":"JLIPHWG3PMFR","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_16","alias_value":"JLIPHWG3PMFRH6SX","created_at":"2026-07-05T11:12:58Z"},{"alias_kind":"pith_short_8","alias_value":"JLIPHWG3","created_at":"2026-07-05T11:12:58Z"}],"graph_snapshots":[{"event_id":"sha256:7455fc5abd3bb6af4a6457f405c246bfdc92f70e53ef957190a0577a16a525a6","target":"graph","created_at":"2026-07-05T11:12:58Z","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/2505.24878/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"CAPTCHAs have been a critical bottleneck for deploying web agents in real-world applications, often blocking them from completing end-to-end automation tasks. While modern multimodal LLM agents have demonstrated impressive performance in static perception tasks, their ability to handle interactive, multi-step reasoning challenges like CAPTCHAs is largely untested. To address this gap, we introduce Open CaptchaWorld, the first web-based benchmark and platform specifically designed to evaluate the visual reasoning and interaction capabilities of MLLM-powered agents through diverse and dynamic CA","authors_text":"Jiacheng Cui, Jiacheng Liu, Xiaohan Zhao, Yaxin Luo, Zhaoyi Li, Zhiqiang Shen","cross_cats":["cs.CL","cs.CV","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T17:59:55Z","title":"Open CaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24878","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":"sha256:7075293f43a26ab5e91e13d20e8475232638616f5a65d24b10582fb79e2f4228","target":"record","created_at":"2026-07-05T11:12:58Z","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":"e200f5d2b475883252c7d8483a7b2670548b01ed5907a2816b8b9bef12b0a702","cross_cats_sorted":["cs.CL","cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T17:59:55Z","title_canon_sha256":"6d2ebec2c6f47a3b49213231b4aa5e6cf70c3cfebea8e12c46d01b9eccf668f9"},"schema_version":"1.0","source":{"id":"2505.24878","kind":"arxiv","version":1}},"canonical_sha256":"4ad0f3d8db7b0b13fa57679a8c2b42c693492f4a131e5984afff4a5606bac8df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4ad0f3d8db7b0b13fa57679a8c2b42c693492f4a131e5984afff4a5606bac8df","first_computed_at":"2026-07-05T11:12:58.228869Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:58.228869Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"76EAjVZMuXtFJA14hLj8hMWAsvluU1jAnc66+nDZ20qMH3xzI/INY0Om/2Zshqr45J8xfPRyztO7wSRf6yaEDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:58.229423Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.24878","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7075293f43a26ab5e91e13d20e8475232638616f5a65d24b10582fb79e2f4228","sha256:7455fc5abd3bb6af4a6457f405c246bfdc92f70e53ef957190a0577a16a525a6"],"state_sha256":"0363a7226d2c1349e9e1c7016cc299b32d09cf92700047ba6100a945f0a1dc00"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zIjndmjpph7jJkc1NQemf59D3K7iD/HHLl5DI+S7/b8bPgAJnUlM/d+ihnav45h/K6i/Qwf7lzTh/91m/o+KBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T21:29:46.930291Z","bundle_sha256":"905652b6a30018b167c24099781c7c18d59c97be5b0031392ae57e4534a16813"}}