{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2J6QV6626IUOJB5FPBDQOMGEI7","short_pith_number":"pith:2J6QV662","schema_version":"1.0","canonical_sha256":"d27d0afbdaf228e487a578470730c447d2f7a5e161b08dcfbaadbe3db7c0af0d","source":{"kind":"arxiv","id":"2606.24525","version":1},"attestation_state":"computed","paper":{"title":"VisCritic: Visual State Comparison as Process Reward for GUI Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiachen Qian","submitted_at":"2026-06-23T12:57:02Z","abstract_excerpt":"GUI agents powered by vision-language models show strong potential for automating digital tasks, yet frequently fail in long-horizon scenarios due to the absence of step-level verification. Existing process reward models verify actions through textual reasoning alone, missing the visual nature of GUI state changes. We introduce VisCritic, a visual process reward framework that verifies agent actions by directly comparing pre-action and post-action screenshots in visual feature space. VisCritic employs a Siamese vision transformer to extract change-aware representations, coupled with an Action-"},"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":"2606.24525","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-23T12:57:02Z","cross_cats_sorted":[],"title_canon_sha256":"041e91814110af02728b00419560201056a36d66715722d0c7e18774dfc5dfa2","abstract_canon_sha256":"4b264ac8b63ad01eea8b3e26ad515dbf6428ab5d956fcc26a31409ce6502f2dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:15:32.944139Z","signature_b64":"AEfV0FNnYE3uH0LdepNVOyrLrEk+7EK5al7HamEIpswG+0aOK25ihxhXb28g0/VKZmh8GAErdTMl5/ik9n+nCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d27d0afbdaf228e487a578470730c447d2f7a5e161b08dcfbaadbe3db7c0af0d","last_reissued_at":"2026-06-24T01:15:32.943815Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:15:32.943815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VisCritic: Visual State Comparison as Process Reward for GUI Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiachen Qian","submitted_at":"2026-06-23T12:57:02Z","abstract_excerpt":"GUI agents powered by vision-language models show strong potential for automating digital tasks, yet frequently fail in long-horizon scenarios due to the absence of step-level verification. Existing process reward models verify actions through textual reasoning alone, missing the visual nature of GUI state changes. We introduce VisCritic, a visual process reward framework that verifies agent actions by directly comparing pre-action and post-action screenshots in visual feature space. VisCritic employs a Siamese vision transformer to extract change-aware representations, coupled with an Action-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24525","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/2606.24525/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":"2606.24525","created_at":"2026-06-24T01:15:32.943868+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24525v1","created_at":"2026-06-24T01:15:32.943868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24525","created_at":"2026-06-24T01:15:32.943868+00:00"},{"alias_kind":"pith_short_12","alias_value":"2J6QV6626IUO","created_at":"2026-06-24T01:15:32.943868+00:00"},{"alias_kind":"pith_short_16","alias_value":"2J6QV6626IUOJB5F","created_at":"2026-06-24T01:15:32.943868+00:00"},{"alias_kind":"pith_short_8","alias_value":"2J6QV662","created_at":"2026-06-24T01:15:32.943868+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/2J6QV6626IUOJB5FPBDQOMGEI7","json":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7.json","graph_json":"https://pith.science/api/pith-number/2J6QV6626IUOJB5FPBDQOMGEI7/graph.json","events_json":"https://pith.science/api/pith-number/2J6QV6626IUOJB5FPBDQOMGEI7/events.json","paper":"https://pith.science/paper/2J6QV662"},"agent_actions":{"view_html":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7","download_json":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7.json","view_paper":"https://pith.science/paper/2J6QV662","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24525&json=true","fetch_graph":"https://pith.science/api/pith-number/2J6QV6626IUOJB5FPBDQOMGEI7/graph.json","fetch_events":"https://pith.science/api/pith-number/2J6QV6626IUOJB5FPBDQOMGEI7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7/action/storage_attestation","attest_author":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7/action/author_attestation","sign_citation":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7/action/citation_signature","submit_replication":"https://pith.science/pith/2J6QV6626IUOJB5FPBDQOMGEI7/action/replication_record"}},"created_at":"2026-06-24T01:15:32.943868+00:00","updated_at":"2026-06-24T01:15:32.943868+00:00"}