{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:JNRUFTLCZD24QNBOJW2OJQ74KP","short_pith_number":"pith:JNRUFTLC","canonical_record":{"source":{"id":"2503.16788","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-03-21T01:52:43Z","cross_cats_sorted":[],"title_canon_sha256":"c30cd6112236db067ed7633cd33c7b5b7b28157349a449f3de71f783fb2ddbc6","abstract_canon_sha256":"c973ddeb1b27498aa33f1c95b126cae40e7647f21a66e7c88c86d419b8512891"},"schema_version":"1.0"},"canonical_sha256":"4b6342cd62c8f5c8342e4db4e4c3fc53e5ba205cd2359880002de5383a4ecc90","source":{"kind":"arxiv","id":"2503.16788","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.16788","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"arxiv_version","alias_value":"2503.16788v1","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.16788","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_12","alias_value":"JNRUFTLCZD24","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_16","alias_value":"JNRUFTLCZD24QNBO","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_8","alias_value":"JNRUFTLC","created_at":"2026-07-05T10:36:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:JNRUFTLCZD24QNBOJW2OJQ74KP","target":"record","payload":{"canonical_record":{"source":{"id":"2503.16788","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-03-21T01:52:43Z","cross_cats_sorted":[],"title_canon_sha256":"c30cd6112236db067ed7633cd33c7b5b7b28157349a449f3de71f783fb2ddbc6","abstract_canon_sha256":"c973ddeb1b27498aa33f1c95b126cae40e7647f21a66e7c88c86d419b8512891"},"schema_version":"1.0"},"canonical_sha256":"4b6342cd62c8f5c8342e4db4e4c3fc53e5ba205cd2359880002de5383a4ecc90","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:49.949587Z","signature_b64":"cWuq0I03KQlBmn8iaxQ7GY7HvAa66itv8LkDzJ36/EuLLLBpFWJ33PhaVeJ7soxE9jwhRhyGoLXTMm+35aufAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b6342cd62c8f5c8342e4db4e4c3fc53e5ba205cd2359880002de5383a4ecc90","last_reissued_at":"2026-07-05T10:36:49.949080Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:49.949080Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.16788","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":"2026-07-05T10:36:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QMjM2m8zKkQu7Hq5/Y85u04bnPSSMO40CPWTLdsA8k0eMVaEST5+eiQ2WXpWEub00funoCVxuS62+3WmmPAKCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T23:28:21.081841Z"},"content_sha256":"29f2e5f119e45ce2c2aa35e48c35c4a8554d1710516748083f91d84e8806056e","schema_version":"1.0","event_id":"sha256:29f2e5f119e45ce2c2aa35e48c35c4a8554d1710516748083f91d84e8806056e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:JNRUFTLCZD24QNBOJW2OJQ74KP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Does Chain-of-Thought Reasoning Help Mobile GUI Agent? An Empirical Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Li Zhang, Longxi Gao, Mengwei Xu","submitted_at":"2025-03-21T01:52:43Z","abstract_excerpt":"Reasoning capabilities have significantly improved the performance of vision-language models (VLMs) in domains such as mathematical problem-solving, coding, and visual question-answering. However, their impact on real-world applications remains unclear. This paper presents the first empirical study on the effectiveness of reasoning-enabled VLMs in mobile GUI agents, a domain that requires interpreting complex screen layouts, understanding user instructions, and executing multi-turn interactions. We evaluate two pairs of commercial models--Gemini 2.0 Flash and Claude 3.7 Sonnet--comparing their"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.16788","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/2503.16788/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":"2026-07-05T10:36:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/zIyaam05SLeAGxIb2q6ugrW32CFXw6eQK8usWquUDoi2FeiUT3MoISgpQoKHVDoXQUV777XZw2BC0OoDBqvCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T23:28:21.082334Z"},"content_sha256":"791ff71fda041b67f73422948e97981706408ef1ff2f6a51ce3eb66c0af39e5e","schema_version":"1.0","event_id":"sha256:791ff71fda041b67f73422948e97981706408ef1ff2f6a51ce3eb66c0af39e5e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/bundle.json","state_url":"https://pith.science/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/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":"2026-08-10T23:28:21Z","links":{"resolver":"https://pith.science/pith/JNRUFTLCZD24QNBOJW2OJQ74KP","bundle":"https://pith.science/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/bundle.json","state":"https://pith.science/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JNRUFTLCZD24QNBOJW2OJQ74KP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:JNRUFTLCZD24QNBOJW2OJQ74KP","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":"c973ddeb1b27498aa33f1c95b126cae40e7647f21a66e7c88c86d419b8512891","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-03-21T01:52:43Z","title_canon_sha256":"c30cd6112236db067ed7633cd33c7b5b7b28157349a449f3de71f783fb2ddbc6"},"schema_version":"1.0","source":{"id":"2503.16788","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.16788","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"arxiv_version","alias_value":"2503.16788v1","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.16788","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_12","alias_value":"JNRUFTLCZD24","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_16","alias_value":"JNRUFTLCZD24QNBO","created_at":"2026-07-05T10:36:49Z"},{"alias_kind":"pith_short_8","alias_value":"JNRUFTLC","created_at":"2026-07-05T10:36:49Z"}],"graph_snapshots":[{"event_id":"sha256:791ff71fda041b67f73422948e97981706408ef1ff2f6a51ce3eb66c0af39e5e","target":"graph","created_at":"2026-07-05T10:36:49Z","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/2503.16788/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reasoning capabilities have significantly improved the performance of vision-language models (VLMs) in domains such as mathematical problem-solving, coding, and visual question-answering. However, their impact on real-world applications remains unclear. This paper presents the first empirical study on the effectiveness of reasoning-enabled VLMs in mobile GUI agents, a domain that requires interpreting complex screen layouts, understanding user instructions, and executing multi-turn interactions. We evaluate two pairs of commercial models--Gemini 2.0 Flash and Claude 3.7 Sonnet--comparing their","authors_text":"Li Zhang, Longxi Gao, Mengwei Xu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-03-21T01:52:43Z","title":"Does Chain-of-Thought Reasoning Help Mobile GUI Agent? An Empirical Study"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.16788","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":"sha256:29f2e5f119e45ce2c2aa35e48c35c4a8554d1710516748083f91d84e8806056e","target":"record","created_at":"2026-07-05T10:36:49Z","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":"c973ddeb1b27498aa33f1c95b126cae40e7647f21a66e7c88c86d419b8512891","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-03-21T01:52:43Z","title_canon_sha256":"c30cd6112236db067ed7633cd33c7b5b7b28157349a449f3de71f783fb2ddbc6"},"schema_version":"1.0","source":{"id":"2503.16788","kind":"arxiv","version":1}},"canonical_sha256":"4b6342cd62c8f5c8342e4db4e4c3fc53e5ba205cd2359880002de5383a4ecc90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4b6342cd62c8f5c8342e4db4e4c3fc53e5ba205cd2359880002de5383a4ecc90","first_computed_at":"2026-07-05T10:36:49.949080Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:36:49.949080Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cWuq0I03KQlBmn8iaxQ7GY7HvAa66itv8LkDzJ36/EuLLLBpFWJ33PhaVeJ7soxE9jwhRhyGoLXTMm+35aufAg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:36:49.949587Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.16788","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:29f2e5f119e45ce2c2aa35e48c35c4a8554d1710516748083f91d84e8806056e","sha256:791ff71fda041b67f73422948e97981706408ef1ff2f6a51ce3eb66c0af39e5e"],"state_sha256":"23dde7d391ec31b88845a4c0a4ac0a392e7a63ecfb9b80543bc3d2c48c496d4a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NuE7uoh5Ti4/EreMZ4MlS7vlFm4GyPVlXBvSyVrz16WwHGstetuebmWIDSGcQv5VmTH8iR9DiPzXaSpl4HoGDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T23:28:21.088503Z","bundle_sha256":"55d4ceede72f055970cf051a5908ad194e17ecd5d735bc015e68720b31f9d52f"}}