{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WS7DCTGE632OGDJXP5SAA3OUVC","short_pith_number":"pith:WS7DCTGE","schema_version":"1.0","canonical_sha256":"b4be314cc4f6f4e30d377f64006dd4a8bbbd960163ff6a421325acc364fc1cf5","source":{"kind":"arxiv","id":"2509.07098","version":1},"attestation_state":"computed","paper":{"title":"Instruction Agent: Enhancing Agent with Expert Demonstration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hailey Hultquist, Justin Wagle, Kazuhito Koishida, Yinheng Li","submitted_at":"2025-09-08T18:00:12Z","abstract_excerpt":"Graphical user interface (GUI) agents have advanced rapidly but still struggle with complex tasks involving novel UI elements, long-horizon actions, and personalized trajectories. In this work, we introduce Instruction Agent, a GUI agent that leverages expert demonstrations to solve such tasks, enabling completion of otherwise difficult workflows. Given a single demonstration, the agent extracts step-by-step instructions and executes them by strictly following the trajectory intended by the user, which avoids making mistakes during execution. The agent leverages the verifier and backtracker mo"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.07098","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-09-08T18:00:12Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"0eb9de417907107ed8aa9491cb14a1178ff5cfa0395ebfaa4f752693cdc6e9e1","abstract_canon_sha256":"1dd38eeb69476adeb102f9f33cff30953e26822eee158789a13fe2c0691958ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:15.543101Z","signature_b64":"lZn7V/yLYTI17qufB8+kTxZeNFS2i/MMG5fe1CzSG6kgpxJUbIgdGv3mouykg5feJtHH5OZ1wJumgTOr8aDlBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4be314cc4f6f4e30d377f64006dd4a8bbbd960163ff6a421325acc364fc1cf5","last_reissued_at":"2026-07-05T12:07:15.542575Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:15.542575Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Instruction Agent: Enhancing Agent with Expert Demonstration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Hailey Hultquist, Justin Wagle, Kazuhito Koishida, Yinheng Li","submitted_at":"2025-09-08T18:00:12Z","abstract_excerpt":"Graphical user interface (GUI) agents have advanced rapidly but still struggle with complex tasks involving novel UI elements, long-horizon actions, and personalized trajectories. In this work, we introduce Instruction Agent, a GUI agent that leverages expert demonstrations to solve such tasks, enabling completion of otherwise difficult workflows. Given a single demonstration, the agent extracts step-by-step instructions and executes them by strictly following the trajectory intended by the user, which avoids making mistakes during execution. The agent leverages the verifier and backtracker mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.07098","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/2509.07098/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.07098","created_at":"2026-07-05T12:07:15.542640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.07098v1","created_at":"2026-07-05T12:07:15.542640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.07098","created_at":"2026-07-05T12:07:15.542640+00:00"},{"alias_kind":"pith_short_12","alias_value":"WS7DCTGE632O","created_at":"2026-07-05T12:07:15.542640+00:00"},{"alias_kind":"pith_short_16","alias_value":"WS7DCTGE632OGDJX","created_at":"2026-07-05T12:07:15.542640+00:00"},{"alias_kind":"pith_short_8","alias_value":"WS7DCTGE","created_at":"2026-07-05T12:07:15.542640+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC","json":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC.json","graph_json":"https://pith.science/api/pith-number/WS7DCTGE632OGDJXP5SAA3OUVC/graph.json","events_json":"https://pith.science/api/pith-number/WS7DCTGE632OGDJXP5SAA3OUVC/events.json","paper":"https://pith.science/paper/WS7DCTGE"},"agent_actions":{"view_html":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC","download_json":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC.json","view_paper":"https://pith.science/paper/WS7DCTGE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.07098&json=true","fetch_graph":"https://pith.science/api/pith-number/WS7DCTGE632OGDJXP5SAA3OUVC/graph.json","fetch_events":"https://pith.science/api/pith-number/WS7DCTGE632OGDJXP5SAA3OUVC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC/action/storage_attestation","attest_author":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC/action/author_attestation","sign_citation":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC/action/citation_signature","submit_replication":"https://pith.science/pith/WS7DCTGE632OGDJXP5SAA3OUVC/action/replication_record"}},"created_at":"2026-07-05T12:07:15.542640+00:00","updated_at":"2026-07-05T12:07:15.542640+00:00"}