{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:F4YLIHERNMVGVJHPOKQXJREDB3","short_pith_number":"pith:F4YLIHER","schema_version":"1.0","canonical_sha256":"2f30b41c916b2a6aa4ef72a174c4830ef9c2db54020fbadec377b814b6bdb4b8","source":{"kind":"arxiv","id":"2605.16116","version":1},"attestation_state":"computed","paper":{"title":"ShopGym: An Integrated Framework for Realistic Simulation and Scalable Benchmarking of E-Commerce Web Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alberto Castelo, Chinmay Savadikar, Han Li, Lingyun Wang, Mingyu Zhao, Shuang Xie, Tianfu Wu, Yuanzheng Zhu","submitted_at":"2026-05-15T16:00:38Z","abstract_excerpt":"Developing and evaluating e-commerce web agents requires environments that preserve meaningful task structure while enabling controllable, reproducible, and scalable scientific comparison. Existing methodologies force a tradeoff: live storefronts provide realism but are non-stationary, difficult to inspect, and irreproducible, while hand-built sandbox benchmarks provide control but cover only a narrow range of layouts, catalogs, policies, and interaction patterns. We argue that the core bottleneck is methodological: the field lacks a scalable way to construct evaluation settings that are simul"},"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":"2605.16116","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-15T16:00:38Z","cross_cats_sorted":[],"title_canon_sha256":"505843657fd92b09ab89fbd80476ad649032f929d0ebe1178f7c4eca29610959","abstract_canon_sha256":"908299c0fed1ee5761f43e8fa519f55646ebb78fd4e2589a2b45a8e9a4c6f041"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:53.561142Z","signature_b64":"RwNYaBJ5a+YaObrK6ZQERMbkgNQAew13/RKWs5x0uyELRHoD1jUUYcYgT6YVsDBGCtaYZ6nwfPUGJ+baEbQMDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f30b41c916b2a6aa4ef72a174c4830ef9c2db54020fbadec377b814b6bdb4b8","last_reissued_at":"2026-05-20T00:01:53.560670Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:53.560670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ShopGym: An Integrated Framework for Realistic Simulation and Scalable Benchmarking of E-Commerce Web Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alberto Castelo, Chinmay Savadikar, Han Li, Lingyun Wang, Mingyu Zhao, Shuang Xie, Tianfu Wu, Yuanzheng Zhu","submitted_at":"2026-05-15T16:00:38Z","abstract_excerpt":"Developing and evaluating e-commerce web agents requires environments that preserve meaningful task structure while enabling controllable, reproducible, and scalable scientific comparison. Existing methodologies force a tradeoff: live storefronts provide realism but are non-stationary, difficult to inspect, and irreproducible, while hand-built sandbox benchmarks provide control but cover only a narrow range of layouts, catalogs, policies, and interaction patterns. We argue that the core bottleneck is methodological: the field lacks a scalable way to construct evaluation settings that are simul"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16116","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/2605.16116/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:33.853892Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T16:41:55.477016Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"1721eceba9e9ec297a81779bcc70fb1a0e63372346ee2fb9e81387be088a57c5"},"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":"2605.16116","created_at":"2026-05-20T00:01:53.560734+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16116v1","created_at":"2026-05-20T00:01:53.560734+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16116","created_at":"2026-05-20T00:01:53.560734+00:00"},{"alias_kind":"pith_short_12","alias_value":"F4YLIHERNMVG","created_at":"2026-05-20T00:01:53.560734+00:00"},{"alias_kind":"pith_short_16","alias_value":"F4YLIHERNMVGVJHP","created_at":"2026-05-20T00:01:53.560734+00:00"},{"alias_kind":"pith_short_8","alias_value":"F4YLIHER","created_at":"2026-05-20T00:01:53.560734+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/F4YLIHERNMVGVJHPOKQXJREDB3","json":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3.json","graph_json":"https://pith.science/api/pith-number/F4YLIHERNMVGVJHPOKQXJREDB3/graph.json","events_json":"https://pith.science/api/pith-number/F4YLIHERNMVGVJHPOKQXJREDB3/events.json","paper":"https://pith.science/paper/F4YLIHER"},"agent_actions":{"view_html":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3","download_json":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3.json","view_paper":"https://pith.science/paper/F4YLIHER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16116&json=true","fetch_graph":"https://pith.science/api/pith-number/F4YLIHERNMVGVJHPOKQXJREDB3/graph.json","fetch_events":"https://pith.science/api/pith-number/F4YLIHERNMVGVJHPOKQXJREDB3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3/action/storage_attestation","attest_author":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3/action/author_attestation","sign_citation":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3/action/citation_signature","submit_replication":"https://pith.science/pith/F4YLIHERNMVGVJHPOKQXJREDB3/action/replication_record"}},"created_at":"2026-05-20T00:01:53.560734+00:00","updated_at":"2026-05-20T00:01:53.560734+00:00"}