{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:CNNZGAOPP2GBQXDBU4MXIRFAKE","short_pith_number":"pith:CNNZGAOP","canonical_record":{"source":{"id":"2508.08645","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-12T05:20:14Z","cross_cats_sorted":[],"title_canon_sha256":"5f252f8562af06d16829c34bdeb0d07f458f1e513034797bec14f03fbf3d222b","abstract_canon_sha256":"205792ea621115e10a185f5f6c0bb570e94a876eafe7035c70408fd5bbff1db0"},"schema_version":"1.0"},"canonical_sha256":"135b9301cf7e8c185c61a7197444a051269d488417dd21c34b87ce6933ae5d24","source":{"kind":"arxiv","id":"2508.08645","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.08645","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"arxiv_version","alias_value":"2508.08645v3","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.08645","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_12","alias_value":"CNNZGAOPP2GB","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_16","alias_value":"CNNZGAOPP2GBQXDB","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_8","alias_value":"CNNZGAOP","created_at":"2026-08-04T00:35:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:CNNZGAOPP2GBQXDBU4MXIRFAKE","target":"record","payload":{"canonical_record":{"source":{"id":"2508.08645","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-12T05:20:14Z","cross_cats_sorted":[],"title_canon_sha256":"5f252f8562af06d16829c34bdeb0d07f458f1e513034797bec14f03fbf3d222b","abstract_canon_sha256":"205792ea621115e10a185f5f6c0bb570e94a876eafe7035c70408fd5bbff1db0"},"schema_version":"1.0"},"canonical_sha256":"135b9301cf7e8c185c61a7197444a051269d488417dd21c34b87ce6933ae5d24","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:35:58.324084Z","signature_b64":"LcIQtXACoQAEelibCykmdCfT8vkN4uINWiismtgMsiqNB6wfqvBQ5v1+4E7ijU2Ezum3HrnBQU9Ydgqy3CrUCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"135b9301cf7e8c185c61a7197444a051269d488417dd21c34b87ce6933ae5d24","last_reissued_at":"2026-08-04T00:35:58.322526Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:35:58.322526Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.08645","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-04T00:35:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g023D47Nzowj6xENGwHOdVLQ2iY6EiC9uAWO1IzTS09sW9p3Y1vA3QFlMyeaSko6hHiC+fj4T5yQ2U26blW4DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T01:05:14.674067Z"},"content_sha256":"a7802648a515fbfcb5b6d5d6d64f192b37d34f57c11cf05b4cfea292f6adab38","schema_version":"1.0","event_id":"sha256:a7802648a515fbfcb5b6d5d6d64f192b37d34f57c11cf05b4cfea292f6adab38"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:CNNZGAOPP2GBQXDBU4MXIRFAKE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Quick on the Uptake: Eliciting Implicit Intents from Human Demonstrations for Personalized Mobile-Use Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Heyuan Huang, Jun Wang, Weinan Zhang, Weiwen Liu, Xingyu Lou, Yanjia Yang, Yuanyi Song, Zheng Wu, Zhuosheng Zhang","submitted_at":"2025-08-12T05:20:14Z","abstract_excerpt":"As multimodal large language models advance rapidly, the automation of mobile tasks has become increasingly feasible through the use of mobile-use agents that mimic human interactions from graphical user interfaces. To further enhance mobile-use agents, previous studies employ demonstration learning to improve mobile-use agents from human demonstrations. However, these methods focus solely on the explicit intention flows of humans (e.g., step sequences) while neglecting implicit intention flows (e.g., personal preferences), which makes it difficult to construct personalized mobile-use agents. "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental results demonstrate that IFRAgent outperforms baselines by an average of 6.79% (32.06% relative improvement) in human intention alignment rate and improves step completion rates by an average of 5.30% (26.34% relative improvement).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The MobileIAR dataset and the implicit intention flow analysis reliably capture stable personal preferences that generalize beyond the collected demonstrations without requiring ongoing user feedback or explicit preference statements.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"IFRAgent builds a query-level SOP library from explicit flows and a user-level habit repository from implicit flows in human demonstrations, then uses SOP extraction, RAG, and query rewriting to raise mobile agent intention alignment by 6.79% on average.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"23319f208e57037106c78a671c09c96eacffe10c89795db03f0459288656658e"},"source":{"id":"2508.08645","kind":"arxiv","version":3},"verdict":{"id":"4a38ad67-4432-4ff1-ad7f-70c90965cb25","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T23:59:33.583537Z","strongest_claim":"Experimental results demonstrate that IFRAgent outperforms baselines by an average of 6.79% (32.06% relative improvement) in human intention alignment rate and improves step completion rates by an average of 5.30% (26.34% relative improvement).","one_line_summary":"IFRAgent builds a query-level SOP library from explicit flows and a user-level habit repository from implicit flows in human demonstrations, then uses SOP extraction, RAG, and query rewriting to raise mobile agent intention alignment by 6.79% on average.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The MobileIAR dataset and the implicit intention flow analysis reliably capture stable personal preferences that generalize beyond the collected demonstrations without requiring ongoing user feedback or explicit preference statements.","pith_extraction_headline":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.08645/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":"4a38ad67-4432-4ff1-ad7f-70c90965cb25"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-04T00:35:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DoFIwYU4jCFbKBGxlHGIpbPwu/kSocItwWD7k9ykD/TeBwatGSRdwWXhQZb/R2k/17UO9+QSiSFaT5Y0pEfRAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T01:05:14.674848Z"},"content_sha256":"68a3d10ecca08d8f456beb4b17961ab7242e3f49c9ae03424d886219a884e74f","schema_version":"1.0","event_id":"sha256:68a3d10ecca08d8f456beb4b17961ab7242e3f49c9ae03424d886219a884e74f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/bundle.json","state_url":"https://pith.science/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/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":"2026-08-05T01:05:14Z","links":{"resolver":"https://pith.science/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE","bundle":"https://pith.science/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/bundle.json","state":"https://pith.science/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CNNZGAOPP2GBQXDBU4MXIRFAKE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CNNZGAOPP2GBQXDBU4MXIRFAKE","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":"205792ea621115e10a185f5f6c0bb570e94a876eafe7035c70408fd5bbff1db0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-12T05:20:14Z","title_canon_sha256":"5f252f8562af06d16829c34bdeb0d07f458f1e513034797bec14f03fbf3d222b"},"schema_version":"1.0","source":{"id":"2508.08645","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.08645","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"arxiv_version","alias_value":"2508.08645v3","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.08645","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_12","alias_value":"CNNZGAOPP2GB","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_16","alias_value":"CNNZGAOPP2GBQXDB","created_at":"2026-08-04T00:35:58Z"},{"alias_kind":"pith_short_8","alias_value":"CNNZGAOP","created_at":"2026-08-04T00:35:58Z"}],"graph_snapshots":[{"event_id":"sha256:68a3d10ecca08d8f456beb4b17961ab7242e3f49c9ae03424d886219a884e74f","target":"graph","created_at":"2026-08-04T00:35:58Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental results demonstrate that IFRAgent outperforms baselines by an average of 6.79% (32.06% relative improvement) in human intention alignment rate and improves step completion rates by an average of 5.30% (26.34% relative improvement)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The MobileIAR dataset and the implicit intention flow analysis reliably capture stable personal preferences that generalize beyond the collected demonstrations without requiring ongoing user feedback or explicit preference statements."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"IFRAgent builds a query-level SOP library from explicit flows and a user-level habit repository from implicit flows in human demonstrations, then uses SOP extraction, RAG, and query rewriting to raise mobile agent intention alignment by 6.79% on average."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents."}],"snapshot_sha256":"23319f208e57037106c78a671c09c96eacffe10c89795db03f0459288656658e"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2508.08645/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As multimodal large language models advance rapidly, the automation of mobile tasks has become increasingly feasible through the use of mobile-use agents that mimic human interactions from graphical user interfaces. To further enhance mobile-use agents, previous studies employ demonstration learning to improve mobile-use agents from human demonstrations. However, these methods focus solely on the explicit intention flows of humans (e.g., step sequences) while neglecting implicit intention flows (e.g., personal preferences), which makes it difficult to construct personalized mobile-use agents. ","authors_text":"Heyuan Huang, Jun Wang, Weinan Zhang, Weiwen Liu, Xingyu Lou, Yanjia Yang, Yuanyi Song, Zheng Wu, Zhuosheng Zhang","cross_cats":[],"headline":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-12T05:20:14Z","title":"Quick on the Uptake: Eliciting Implicit Intents from Human Demonstrations for Personalized Mobile-Use Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.08645","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-18T23:59:33.583537Z","id":"4a38ad67-4432-4ff1-ad7f-70c90965cb25","model_set":{"reader":"grok-4.3"},"one_line_summary":"IFRAgent builds a query-level SOP library from explicit flows and a user-level habit repository from implicit flows in human demonstrations, then uses SOP extraction, RAG, and query rewriting to raise mobile agent intention alignment by 6.79% on average.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"IFRAgent extracts implicit user preferences from demonstrations to build personalized habit repositories for mobile agents.","strongest_claim":"Experimental results demonstrate that IFRAgent outperforms baselines by an average of 6.79% (32.06% relative improvement) in human intention alignment rate and improves step completion rates by an average of 5.30% (26.34% relative improvement).","weakest_assumption":"The MobileIAR dataset and the implicit intention flow analysis reliably capture stable personal preferences that generalize beyond the collected demonstrations without requiring ongoing user feedback or explicit preference statements."}},"verdict_id":"4a38ad67-4432-4ff1-ad7f-70c90965cb25"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a7802648a515fbfcb5b6d5d6d64f192b37d34f57c11cf05b4cfea292f6adab38","target":"record","created_at":"2026-08-04T00:35:58Z","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":"205792ea621115e10a185f5f6c0bb570e94a876eafe7035c70408fd5bbff1db0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-12T05:20:14Z","title_canon_sha256":"5f252f8562af06d16829c34bdeb0d07f458f1e513034797bec14f03fbf3d222b"},"schema_version":"1.0","source":{"id":"2508.08645","kind":"arxiv","version":3}},"canonical_sha256":"135b9301cf7e8c185c61a7197444a051269d488417dd21c34b87ce6933ae5d24","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"135b9301cf7e8c185c61a7197444a051269d488417dd21c34b87ce6933ae5d24","first_computed_at":"2026-08-04T00:35:58.322526Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T00:35:58.322526Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LcIQtXACoQAEelibCykmdCfT8vkN4uINWiismtgMsiqNB6wfqvBQ5v1+4E7ijU2Ezum3HrnBQU9Ydgqy3CrUCw==","signature_status":"signed_v1","signed_at":"2026-08-04T00:35:58.324084Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.08645","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a7802648a515fbfcb5b6d5d6d64f192b37d34f57c11cf05b4cfea292f6adab38","sha256:68a3d10ecca08d8f456beb4b17961ab7242e3f49c9ae03424d886219a884e74f"],"state_sha256":"0eb1607be910335c6284a98c463b8d54f037f20f9fdefd62e931ff25dd9f9c3e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"afUZH8tRU+UI62PUiRaex9foKZbfGw90UNwgK9qTohVJMq0y66myECjr8ndPq31jWRcttm88Af7VUMFGyasPBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T01:05:14.678852Z","bundle_sha256":"d6a91f2d22604d8cc3bcc230918a1e801d74b03c2755332cb26f813f598010c7"}}