{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NK5XGTAL6NH6X4WFTAB6DVT5D7","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":"c2fb0a7814d6901a129b355d5f051d87735e2a8a67845dafade2b3a13046d10a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T16:01:38Z","title_canon_sha256":"e7f2ff19b1e4319c0245d9911d71a3b48c5fcb531179a8ef20dc1b1b01811a0d"},"schema_version":"1.0","source":{"id":"2607.25904","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.25904","created_at":"2026-07-29T01:26:10Z"},{"alias_kind":"arxiv_version","alias_value":"2607.25904v1","created_at":"2026-07-29T01:26:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.25904","created_at":"2026-07-29T01:26:10Z"},{"alias_kind":"pith_short_12","alias_value":"NK5XGTAL6NH6","created_at":"2026-07-29T01:26:10Z"},{"alias_kind":"pith_short_16","alias_value":"NK5XGTAL6NH6X4WF","created_at":"2026-07-29T01:26:10Z"},{"alias_kind":"pith_short_8","alias_value":"NK5XGTAL","created_at":"2026-07-29T01:26:10Z"}],"graph_snapshots":[{"event_id":"sha256:985377f139757386a451e34622a6ae2ee395ec45143ca89386c8525e06695d2f","target":"graph","created_at":"2026-07-29T01:26:10Z","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/2607.25904/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Graphical user interface task evaluation aims to determine whether a GUI agent has successfully completed a user instruction. Automated GUI task evaluation has received increasing attention because the evaluation results can serve as reward signals for both test-time scaling and post-training. However, reliable GUI task evaluation remains challenging because the judgments often require access to environment states, such as system configurations, file data, and application settings, beyond the screenshots of execution trajectories. In this paper, we propose an interactive reward agent (IRA) bas","authors_text":"Chenrui Shi, Che Sun, Lifeng Fan, Ruining Feng, Yang Liu, Yuwei Wu, Zhi Gao, Zirui Shang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T16:01:38Z","title":"Interactive Reward Agent: GUI Task Evaluation via Environment-State Verification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.25904","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:e21eebe043993322b410ab8288cb734bf454d898cfcb1fa43b985d67978c886f","target":"record","created_at":"2026-07-29T01:26:10Z","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":"c2fb0a7814d6901a129b355d5f051d87735e2a8a67845dafade2b3a13046d10a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-28T16:01:38Z","title_canon_sha256":"e7f2ff19b1e4319c0245d9911d71a3b48c5fcb531179a8ef20dc1b1b01811a0d"},"schema_version":"1.0","source":{"id":"2607.25904","kind":"arxiv","version":1}},"canonical_sha256":"6abb734c0bf34febf2c59803e1d67d1fccd9b82a0d71c8290077c31e91cf8d66","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6abb734c0bf34febf2c59803e1d67d1fccd9b82a0d71c8290077c31e91cf8d66","first_computed_at":"2026-07-29T01:26:10.301556Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-29T01:26:10.301556Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VGPqb1azBBodJsx1FsROyLbzPtRcoj8U1C2NSt5gmz2v8dU2QR8/ra8QpTKrv6iip51K2iTBKOACm0SpHkdyBQ==","signature_status":"signed_v1","signed_at":"2026-07-29T01:26:10.302374Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.25904","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e21eebe043993322b410ab8288cb734bf454d898cfcb1fa43b985d67978c886f","sha256:985377f139757386a451e34622a6ae2ee395ec45143ca89386c8525e06695d2f"],"state_sha256":"c38711a430035b2c3fd039c54807ba8bd347c83bef63fd393c3cf49f7b4b8737"}