{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.27805", "kind": "paper", "url": "https://arxiv.org/abs/2605.27805"}, {"id": "gh:ziyanluo/childeval", "kind": "repo", "url": "https://github.com/ziyanluo/ChildEval"}], "authors": ["Yanyan Luo", "Xue Han", "Chunxu Zhao", "Ruiqiao Bai", "Yaxing Zhang", "Qian Hu", "Lijun Mei", "Junlan Feng"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.27805", "source_url": "https://arxiv.org/abs/2605.27805"}, "categories": ["General AI", "Language & Knowledge", "Long Context"], "description": {"en": "ChildEval is a benchmark for evaluating LLMs' ability to infer and follow child-centered preferences in long-context conversations. It contains 29K synthesized persona profiles of children aged 3-6, with explicit and implicit preference expressions across five top-level and fourteen sub-level categories."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.27805", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.27805", "title": "ChildEval"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["China Mobile (China)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2605.27805", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2605.27805", "source_url": "https://openalex.org/W7162653081"}, "key": "claire-radar:2605.27805", "modality": null, "name": "ChildEval", "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_childeval_9775399e", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "2c8992c690c1f2615dc96b401490a63a2c9de935ebcd61d3a8a3f711de596408", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.27805"}, "publisher": null, "released": "2026-05-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.27805", "source_url": "https://arxiv.org/abs/2605.27805"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-27805", "source": "claire_radar", "source_benchmark_id": "2605.27805", "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": "https://github.com/ziyanluo/ChildEval", "githubScope": "benchmark_repo", "githubStars": 1, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2605.27805", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "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": "17dc3cabe5cc52157285d91e37e47bd5f0faaa2b978a936491bf9f3f1c02adc8", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ChildEval is a benchmark for evaluating LLMs' ability to infer and follow child-centered preferences in long-context conversations. It contains 29K synthesized persona profiles of children aged 3-6, with explicit and implicit preference expressions across five top-level and fourteen sub-level categories.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "To address this gap, we introduce ChildEval, a benchmark for evaluating LLMs' ability to infer and follow child-centered preferences in long-context conversations."}, "familyId": "bmf_61cce4f98ffa", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/ziyanluo/ChildEval", "url": "https://github.com/ziyanluo/ChildEval"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/ziyanluo/ChildEval", "url": "https://huggingface.co/papers/2605.27805"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/ziyanluo/ChildEval.", "url": "https://arxiv.org/abs/2605.27805"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/ziyanluo/ChildEval", "url": "https://arxiv.org/abs/2605.27805"}], "scope": "benchmark_repo", "url": "https://github.com/ziyanluo/ChildEval", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_childeval_9775399e", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/ziyanluo/ChildEval", "data": null, "hfPaper": "https://huggingface.co/papers/2605.27805", "paper": "https://arxiv.org/abs/2605.27805", "pdf": "https://arxiv.org/pdf/2605.27805", "project": null, "report": "https://arxiv.org/abs/2605.27805"}, "metrics": [], "motivation": "While LLMs enable personalized chatbots, their effectiveness in child-centered personalization remains unclear, as systematic evaluation of child-specific preferences is still lacking.", "name": "ChildEval", "oneLine": "ChildEval is a benchmark for evaluating LLMs' ability to infer and follow child-centered preferences in long-context conversations. It contains 29K synthesized persona profiles of children aged 3-6, with explicit and implicit preference expressions across five top-level and fourteen sub-level categories.", "paperTitle": "ChildEval: When large language models meet children's personalities", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.27805", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "ChildEval Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/ziyanluo/ChildEval"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-27", "source": {"authors": ["Yanyan Luo", "Xue Han", "Chunxu Zhao", "Ruiqiao Bai", "Yaxing Zhang", "Qian Hu", "Lijun Mei", "Junlan Feng"], "categories": ["cs.CL", "cs.AI"], "id": "2605.27805", "title": "ChildEval: When large language models meet children's personalities", "type": "arxiv", "url": "https://arxiv.org/abs/2605.27805"}, "sourceUpdatedAt": "2026-05-27T00:53:42Z", "topics": ["Long Context"], "venueAttempts": [], "whyItMatters": "Personalization for children is under-explored relative to adults. ChildEval provides a protocol to test whether LLMs can infer and follow child-specific preferences, addressing a gap in personalized conversational AI evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
