{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AMXCYM4XERMWOEHY7AHO7O6EF5","short_pith_number":"pith:AMXCYM4X","canonical_record":{"source":{"id":"2501.02863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-06T09:10:11Z","cross_cats_sorted":[],"title_canon_sha256":"d0df0e66be701027d0276e427aba3a2913d2aff8a63a259b2a6ee13c12b41930","abstract_canon_sha256":"59ec8b6e78117463a8524324a4a4b5b5fcbc8740445333818683857cc3acd51b"},"schema_version":"1.0"},"canonical_sha256":"032e2c339724596710f8f80eefbbc42f703d2b5b5e377a1e546b048aa56b98f4","source":{"kind":"arxiv","id":"2501.02863","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.02863","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"arxiv_version","alias_value":"2501.02863v2","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02863","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_12","alias_value":"AMXCYM4XERMW","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_16","alias_value":"AMXCYM4XERMWOEHY","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_8","alias_value":"AMXCYM4X","created_at":"2026-07-05T10:12:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AMXCYM4XERMWOEHY7AHO7O6EF5","target":"record","payload":{"canonical_record":{"source":{"id":"2501.02863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-06T09:10:11Z","cross_cats_sorted":[],"title_canon_sha256":"d0df0e66be701027d0276e427aba3a2913d2aff8a63a259b2a6ee13c12b41930","abstract_canon_sha256":"59ec8b6e78117463a8524324a4a4b5b5fcbc8740445333818683857cc3acd51b"},"schema_version":"1.0"},"canonical_sha256":"032e2c339724596710f8f80eefbbc42f703d2b5b5e377a1e546b048aa56b98f4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:24.952093Z","signature_b64":"6VwO/aYZLBx6E96heRA9J2ekpTfotKvCNnWoTTZZcoZ1J1W8tkbw2/znsN1MzCxdQwj2LK2tUkfczfwn3TJ5BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"032e2c339724596710f8f80eefbbc42f703d2b5b5e377a1e546b048aa56b98f4","last_reissued_at":"2026-07-05T10:12:24.951621Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:24.951621Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.02863","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:12:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"x88xdJmtrInapdti+x6PYEIxr0hzPYzYUtBhxeWwywlr3a1YdmABRe2k2o411xyfD4Lj2tr4c3lf6zsyPTXPAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:34:14.581251Z"},"content_sha256":"805a566b6128936ffbabbf3442b339c78d8deeb19500cffaffea15f2bc5c3a7b","schema_version":"1.0","event_id":"sha256:805a566b6128936ffbabbf3442b339c78d8deeb19500cffaffea15f2bc5c3a7b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AMXCYM4XERMWOEHY7AHO7O6EF5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Assaf Marron, David Harel, Dezhi Ran, Haochuan Lu, Hao Yu, Jun Ren, Liangchao Yao, Mengqian Xu, Mengzhou Wu, Tao Xie, Ting Su, Ting Xiong, Wei Yang, Xia Zeng, Yuan Cao, Yuetang Deng, Yuetong Li, Zexin Xu","submitted_at":"2025-01-06T09:10:11Z","abstract_excerpt":"Recent advances of foundation models (FMs) have made navigating mobile applications (apps) based on high-level goal instructions within reach, with significant industrial applications such as UI testing. While existing benchmarks evaluate FM-based UI navigation using the binary pass/fail metric, they have two major limitations: they cannot reflect the complex nature of mobile UI navigation where FMs may fail for various reasons (e.g., misunderstanding instructions and failed planning), and they lack industrial relevance due to oversimplified tasks that poorly represent real-world scenarios. To"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02863","kind":"arxiv","version":2},"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/2501.02863/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":"2026-07-05T10:12:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OPVl4Xg0A325tpOT8DxLdMOZYl0mivR6btvUEp3iyCjLaFRCW/II+zlftTXBaw3gkr3NRLE79+nz+N1juMKDCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:34:14.581650Z"},"content_sha256":"45c8cccd65c13564286e98dcad2c530c5473d4d35f7b271b4fde6bc71aa8dba9","schema_version":"1.0","event_id":"sha256:45c8cccd65c13564286e98dcad2c530c5473d4d35f7b271b4fde6bc71aa8dba9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/bundle.json","state_url":"https://pith.science/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/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":"2026-08-05T16:34:14Z","links":{"resolver":"https://pith.science/pith/AMXCYM4XERMWOEHY7AHO7O6EF5","bundle":"https://pith.science/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/bundle.json","state":"https://pith.science/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AMXCYM4XERMWOEHY7AHO7O6EF5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AMXCYM4XERMWOEHY7AHO7O6EF5","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":"59ec8b6e78117463a8524324a4a4b5b5fcbc8740445333818683857cc3acd51b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-06T09:10:11Z","title_canon_sha256":"d0df0e66be701027d0276e427aba3a2913d2aff8a63a259b2a6ee13c12b41930"},"schema_version":"1.0","source":{"id":"2501.02863","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.02863","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"arxiv_version","alias_value":"2501.02863v2","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02863","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_12","alias_value":"AMXCYM4XERMW","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_16","alias_value":"AMXCYM4XERMWOEHY","created_at":"2026-07-05T10:12:24Z"},{"alias_kind":"pith_short_8","alias_value":"AMXCYM4X","created_at":"2026-07-05T10:12:24Z"}],"graph_snapshots":[{"event_id":"sha256:45c8cccd65c13564286e98dcad2c530c5473d4d35f7b271b4fde6bc71aa8dba9","target":"graph","created_at":"2026-07-05T10:12:24Z","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/2501.02863/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances of foundation models (FMs) have made navigating mobile applications (apps) based on high-level goal instructions within reach, with significant industrial applications such as UI testing. While existing benchmarks evaluate FM-based UI navigation using the binary pass/fail metric, they have two major limitations: they cannot reflect the complex nature of mobile UI navigation where FMs may fail for various reasons (e.g., misunderstanding instructions and failed planning), and they lack industrial relevance due to oversimplified tasks that poorly represent real-world scenarios. To","authors_text":"Assaf Marron, David Harel, Dezhi Ran, Haochuan Lu, Hao Yu, Jun Ren, Liangchao Yao, Mengqian Xu, Mengzhou Wu, Tao Xie, Ting Su, Ting Xiong, Wei Yang, Xia Zeng, Yuan Cao, Yuetang Deng, Yuetong Li, Zexin Xu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-06T09:10:11Z","title":"Beyond Pass or Fail: Multi-Dimensional Benchmarking of Foundation Models for Goal-based Mobile UI Navigation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02863","kind":"arxiv","version":2},"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":"sha256:805a566b6128936ffbabbf3442b339c78d8deeb19500cffaffea15f2bc5c3a7b","target":"record","created_at":"2026-07-05T10:12:24Z","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":"59ec8b6e78117463a8524324a4a4b5b5fcbc8740445333818683857cc3acd51b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2025-01-06T09:10:11Z","title_canon_sha256":"d0df0e66be701027d0276e427aba3a2913d2aff8a63a259b2a6ee13c12b41930"},"schema_version":"1.0","source":{"id":"2501.02863","kind":"arxiv","version":2}},"canonical_sha256":"032e2c339724596710f8f80eefbbc42f703d2b5b5e377a1e546b048aa56b98f4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"032e2c339724596710f8f80eefbbc42f703d2b5b5e377a1e546b048aa56b98f4","first_computed_at":"2026-07-05T10:12:24.951621Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:12:24.951621Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6VwO/aYZLBx6E96heRA9J2ekpTfotKvCNnWoTTZZcoZ1J1W8tkbw2/znsN1MzCxdQwj2LK2tUkfczfwn3TJ5BA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:12:24.952093Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.02863","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:805a566b6128936ffbabbf3442b339c78d8deeb19500cffaffea15f2bc5c3a7b","sha256:45c8cccd65c13564286e98dcad2c530c5473d4d35f7b271b4fde6bc71aa8dba9"],"state_sha256":"e738aeed7c8384ff933ec150a6391a57a2c962e1a684c08426142d2cdea92bae"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3VByt/lnYCDuySLS72AZ/4dev8Lp6LyI/c5Bh86VKglmXvB+URURnLiZWdEomqtAieeYsATWhWXSxoUmmwzMAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T16:34:14.584001Z","bundle_sha256":"5d8e1ba6064dd0b45803e18708031e28f1a0112acf3fb2699518374cac118ee6"}}