{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ZCR7OSTYPQXNAV2ZLZ36OGWUH2","short_pith_number":"pith:ZCR7OSTY","schema_version":"1.0","canonical_sha256":"c8a3f74a787c2ed057595e77e71ad43e857e12b5a772d1dcd1961f89d4b43a35","source":{"kind":"arxiv","id":"2110.08653","version":1},"attestation_state":"computed","paper":{"title":"Learning UI Navigation through Demonstrations composed of Macro Actions","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Wei Li","submitted_at":"2021-10-16T20:29:41Z","abstract_excerpt":"We have developed a framework to reliably build agents capable of UI navigation. The state space is simplified from raw-pixels to a set of UI elements extracted from screen understanding, such as OCR and icon detection. The action space is restricted to the UI elements plus a few global actions. Actions can be customized for tasks and each action is a sequence of basic operations conditioned on status checks. With such a design, we are able to train DQfD and BC agents with a small number of demonstration episodes. We propose demo augmentation that significantly reduces the required number of h"},"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":"2110.08653","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2021-10-16T20:29:41Z","cross_cats_sorted":[],"title_canon_sha256":"3da7c5de05b16577229746fd05c27ad560770a54d04a0b1c452de14a4b63aaef","abstract_canon_sha256":"c1fcbccd2eeca61df5d177344b32d0b1eabd32d8d2e1a8a5519034d66f0eb882"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:23:18.949739Z","signature_b64":"CCjLRBQqEnHPhCaPOtGO3pRKEHheQFRSDqL+ftNmomsUReGdzw7kVcdd54jELOCKygWR3yapbnY+46aN7bLiDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8a3f74a787c2ed057595e77e71ad43e857e12b5a772d1dcd1961f89d4b43a35","last_reissued_at":"2026-07-05T03:23:18.949334Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:23:18.949334Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning UI Navigation through Demonstrations composed of Macro Actions","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Wei Li","submitted_at":"2021-10-16T20:29:41Z","abstract_excerpt":"We have developed a framework to reliably build agents capable of UI navigation. The state space is simplified from raw-pixels to a set of UI elements extracted from screen understanding, such as OCR and icon detection. The action space is restricted to the UI elements plus a few global actions. Actions can be customized for tasks and each action is a sequence of basic operations conditioned on status checks. With such a design, we are able to train DQfD and BC agents with a small number of demonstration episodes. We propose demo augmentation that significantly reduces the required number of h"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.08653","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/2110.08653/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":"2110.08653","created_at":"2026-07-05T03:23:18.949389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.08653v1","created_at":"2026-07-05T03:23:18.949389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.08653","created_at":"2026-07-05T03:23:18.949389+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZCR7OSTYPQXN","created_at":"2026-07-05T03:23:18.949389+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZCR7OSTYPQXNAV2Z","created_at":"2026-07-05T03:23:18.949389+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZCR7OSTY","created_at":"2026-07-05T03:23:18.949389+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2501.16150","citing_title":"A Comprehensive Survey of Agents for Computer Use: Foundations, Challenges, and Future Directions","ref_index":86,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2","json":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2.json","graph_json":"https://pith.science/api/pith-number/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/graph.json","events_json":"https://pith.science/api/pith-number/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/events.json","paper":"https://pith.science/paper/ZCR7OSTY"},"agent_actions":{"view_html":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2","download_json":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2.json","view_paper":"https://pith.science/paper/ZCR7OSTY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.08653&json=true","fetch_graph":"https://pith.science/api/pith-number/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/graph.json","fetch_events":"https://pith.science/api/pith-number/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/action/storage_attestation","attest_author":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/action/author_attestation","sign_citation":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/action/citation_signature","submit_replication":"https://pith.science/pith/ZCR7OSTYPQXNAV2ZLZ36OGWUH2/action/replication_record"}},"created_at":"2026-07-05T03:23:18.949389+00:00","updated_at":"2026-07-05T03:23:18.949389+00:00"}