{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ROKB2YFQQOCFMEUEVG2A7KSYQF","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":"626de3152150a606d8a60c0f11196e3c670b437639f0ead885deed3d607e5b4b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-06T09:51:30Z","title_canon_sha256":"1d066b2fc399e6706edb5bce7e574a5eb34d5f66d4937d115b271e51c802c488"},"schema_version":"1.0","source":{"id":"2508.04266","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.04266","created_at":"2026-06-19T16:12:47Z"},{"alias_kind":"arxiv_version","alias_value":"2508.04266v4","created_at":"2026-06-19T16:12:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.04266","created_at":"2026-06-19T16:12:47Z"},{"alias_kind":"pith_short_12","alias_value":"ROKB2YFQQOCF","created_at":"2026-06-19T16:12:47Z"},{"alias_kind":"pith_short_16","alias_value":"ROKB2YFQQOCFMEUE","created_at":"2026-06-19T16:12:47Z"},{"alias_kind":"pith_short_8","alias_value":"ROKB2YFQ","created_at":"2026-06-19T16:12:47Z"}],"graph_snapshots":[{"event_id":"sha256:9781cca2f71f0178c63bba730a724c850baab1ecac3c22da616eb2b4fc966987","target":"graph","created_at":"2026-06-19T16:12:47Z","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/2508.04266/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing benchmarks in e-commerce primarily focus on basic user intents, such as finding or purchasing products. However, real-world users often pursue more complex goals, such as applying vouchers, managing budgets, and finding multi-products seller. To bridge this gap, we propose ShoppingBench, a novel end-to-end shopping benchmark designed to encompass increasingly challenging levels of grounded intent. Specifically, we propose a scalable framework to simulate user instructions based on various intents derived from sampled real-world products. To facilitate consistent and reliable evaluatio","authors_text":"Huaipeng Zhao, Jian Dong Zhang, Jiangyuan Wang, Kejun Xiao, Qi Sun, Tao Luo, Xiaoyi Zeng","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-06T09:51:30Z","title":"ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.04266","kind":"arxiv","version":4},"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:ed66ca2b87b2f485c184581f6525a1c072fe655334dbfbe5fb4b1b44005bc013","target":"record","created_at":"2026-06-19T16:12:47Z","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":"626de3152150a606d8a60c0f11196e3c670b437639f0ead885deed3d607e5b4b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-06T09:51:30Z","title_canon_sha256":"1d066b2fc399e6706edb5bce7e574a5eb34d5f66d4937d115b271e51c802c488"},"schema_version":"1.0","source":{"id":"2508.04266","kind":"arxiv","version":4}},"canonical_sha256":"8b941d60b08384561284a9b40faa58815b61525aeadb99765151142e8f47f49a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8b941d60b08384561284a9b40faa58815b61525aeadb99765151142e8f47f49a","first_computed_at":"2026-06-19T16:12:47.128968Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:12:47.128968Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2sdVS1Qy2wGZR6wdXdyB1JPjtlAubpra9ObRLiGLtRWf60vdVhcbCaCrz8SzEJk/QoedZxbDFRquuottNWg3BQ==","signature_status":"signed_v1","signed_at":"2026-06-19T16:12:47.129454Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.04266","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ed66ca2b87b2f485c184581f6525a1c072fe655334dbfbe5fb4b1b44005bc013","sha256:9781cca2f71f0178c63bba730a724c850baab1ecac3c22da616eb2b4fc966987"],"state_sha256":"0355edafcc2701fbf0ca05f81e6bb1d0e11ea087c1bd3d6495bf3c6280d1c478"}