{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.17698", "kind": "paper", "url": "https://arxiv.org/abs/2606.17698"}], "authors": ["Zeyao Du", "Tong Li", "Haibo Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.17698", "source_url": "https://arxiv.org/abs/2606.17698"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "EComAgentBench evaluates LLM-based shopping agents on 662 long-horizon product selection tasks built from real Amazon data. Each task requires uncovering hidden requirements spread across an explicit query, tool-gated profile, and scripted clarification, verifying candidates, and committing to one product within 100 tool calls. Scoring uses typed, source-tagged rubrics for each requirement."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.17698", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.17698", "title": "EComAgentBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.17698", "modality": null, "name": "EComAgentBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_ecomagentbench_d7694649", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "a6fbaf2a36d20a809cbc263447bcf7929cae4c1104d45fcac7b09fd656167902", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.17698"}, "publisher": null, "released": "2026-06-16", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.17698", "source_url": "https://arxiv.org/abs/2606.17698"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-17698", "source": "claire_radar", "source_benchmark_id": "2606.17698", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.17698", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-02", "state": "stale"}}, "status": "stale"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "c4d3fc0fd3c1488e9eda11c19090312cb3837eb2a1ce41843f9c0eb38eed9261", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "EComAgentBench evaluates LLM-based shopping agents on 662 long-horizon product selection tasks built from real Amazon data. Each task requires uncovering hidden requirements spread across an explicit query, tool-gated profile, and scripted clarification, verifying candidates, and committing to one product within 100 tool calls. Scoring uses typed, source-tagged rubrics for each requirement.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To address this gap, we introduce EComAgentBench, a benchmark of 662 tasks grounded in real Amazon products and reviews."}, "familyId": "bmf_7733964f2cda", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_ecomagentbench_d7694649", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.17698", "paper": "https://arxiv.org/abs/2606.17698", "pdf": "https://arxiv.org/pdf/2606.17698", "project": null, "report": "https://arxiv.org/abs/2606.17698"}, "metrics": [], "motivation": "As LLM-based shopping agents enter production, existing benchmarks fail to capture how a shopper's requirements arrive: stated implicitly in the query, recorded in a profile, or revealed only when the right question is asked.", "name": "EComAgentBench", "oneLine": "EComAgentBench evaluates LLM-based shopping agents on 662 long-horizon product selection tasks built from real Amazon data. Each task requires uncovering hidden requirements spread across an explicit query, tool-gated profile, and scripted clarification, verifying candidates, and committing to one product within 100 tool calls. Scoring uses typed, source-tagged rubrics for each requirement.", "paperTitle": "EComAgentBench: Benchmarking Shopping Agents on Long-Horizon Tasks with Distributed Hidden Intent", "primaryDomain": "General AI", "publication": {"evidence": "16 pages, 4 figures; accepted to the EMNLP 2026 Industry Track (camera-ready version)", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.17698", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "EMNLP 2026 Industry Track (camera-ready version)", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-16", "source": {"authors": ["Zeyao Du", "Tong Li", "Yanci Zhang", "Haibo Zhang"], "categories": ["cs.AI", "cs.CL"], "id": "2606.17698", "title": "EComAgentBench: Benchmarking Shopping Agents on Long-Horizon Tasks with Distributed Hidden Intent", "type": "arxiv", "url": "https://arxiv.org/abs/2606.17698"}, "sourceUpdatedAt": "2026-06-23T08:51:01Z", "topics": ["cs.AI"], "venueAttempts": [{"decisionRaw": "16 pages, 4 figures; accepted to the EMNLP 2026 Industry Track (camera-ready version)", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "16 pages, 4 figures; accepted to the EMNLP 2026 Industry Track (camera-ready version)", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.17698"}], "reviewStatus": "accepted", "venueName": "EMNLP 2026 Industry Track (camera-ready version)"}], "whyItMatters": "Existing shopping benchmarks reveal full intent upfront, failing to reflect real-world requirements that emerge over time. EComAgentBench measures agents' ability to handle long-horizon interactions, providing a reproducible foundation for comparing dependable shopping assistance."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
