{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.08160", "kind": "paper", "url": "https://arxiv.org/abs/2608.08160"}, {"id": "gh:yingpengma/ncp-bench", "kind": "repo", "url": "https://github.com/yingpengma/NCP-Bench"}], "authors": ["Yingpeng Ma", "Jianhao Yan", "Bei Shi", "Ka Hou Kam", "Runnan Wang", "Xuebo Liu", "Yulong Chen", "Yue Zhang", "Derek F. Wong"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.08160", "source_url": "https://arxiv.org/abs/2608.08160"}, "categories": ["Agents & Tool Use", "General AI", "cs.CL"], "description": {"en": "NCP-Bench evaluates narrative commitment preservation in interactive narratives. It provides 100 movie-synopsis-based environments with structured narrative specifications and an automatic evaluator that checks fact, commitment, and trajectory consistency across narrator responses to player actions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.08160", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.08160", "title": "NCP-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["National Library of the Philippines", "University of Macau", "Westlake University", "Harbin Institute of Technology", "University of Aberdeen", "University of Cambridge"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.08160", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.08160", "source_url": "https://openalex.org/W7202171566"}, "key": "claire-radar:2608.08160", "modality": null, "name": "NCP-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "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_ncp-bench_83c91412", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "2f4bb5df009012884844f47c9b933ffb46fc954279921c91ce2ace8864a54629", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.08160"}, "publisher": null, "released": "2026-08-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.08160", "source_url": "https://arxiv.org/abs/2608.08160"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-08160", "source": "claire_radar", "source_benchmark_id": "2608.08160", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Agents & Tool Use", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/yingpengma/NCP-Bench", "githubScope": "benchmark_repo", "githubStars": 34, "hfDailySubmittedAt": "2026-08-13T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 30, "hfPaperUrl": "https://huggingface.co/papers/2608.08160", "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": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "Medium", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "61b0d5cc322af2e5dcd34aabfcf5d9257813a42e0dbee2d2254ad8a895921b72", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "NCP-Bench evaluates narrative commitment preservation in interactive narratives. It provides 100 movie-synopsis-based environments with structured narrative specifications and an automatic evaluator that checks fact, commitment, and trajectory consistency across narrator responses to player actions.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence"], "snippet": "We introduce NCP-Bench, a benchmark of 100 narrative environments derived from movie synopses."}, "familyId": "bmf_b98627ff590e", "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/yingpengma/NCP-Bench", "url": "https://github.com/yingpengma/NCP-Bench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/yingpengma/NCP-Bench", "url": "https://huggingface.co/papers/2608.08160"}], "scope": "benchmark_repo", "url": "https://github.com/yingpengma/NCP-Bench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_ncp-bench_83c91412", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/yingpengma/NCP-Bench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.08160", "paper": "https://arxiv.org/abs/2608.08160", "pdf": "https://arxiv.org/pdf/2608.08160", "project": null, "report": "https://arxiv.org/abs/2608.08160"}, "metrics": [], "motivation": "The rapid advancement of Large Language Models (LLMs) is revolutionizing AI for Games by enabling open-ended and fluid interactive storytelling.", "name": "NCP-Bench", "oneLine": "NCP-Bench evaluates narrative commitment preservation in interactive narratives. It provides 100 movie-synopsis-based environments with structured narrative specifications and an automatic evaluator that checks fact, commitment, and trajectory consistency across narrator responses to player actions.", "paperTitle": "Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives", "primaryDomain": "General AI", "publication": {"evidence": "Accepted by ICML 2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2608.08160", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "ICML 2026", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.47457505484152435, "value": 34}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.606767028072414, "value": 30}}, "confidence": "Medium", "coverage": 0.7, "rank": 62, "score": 50}, "method": "attention-ranking-v13", "rank": 62, "score": 50}}, "readiness": "Runnable", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-08-08", "source": {"authors": ["Yingpeng Ma", "Jianhao Yan", "Bei Shi", "Ka Hou Kam", "Runnan Wang", "Xuebo Liu", "Yulong Chen", "Yue Zhang", "Derek F. Wong"], "categories": ["cs.CL", "cs.AI"], "id": "2608.08160", "title": "Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives", "type": "arxiv", "url": "https://arxiv.org/abs/2608.08160"}, "sourceUpdatedAt": "2026-08-08T14:38:18Z", "topics": ["cs.CL"], "venueAttempts": [{"decisionRaw": "Accepted by ICML 2026", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted by ICML 2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2608.08160"}], "reviewStatus": "accepted", "venueName": "ICML 2026"}], "whyItMatters": "The benchmark fills a gap in evaluating long-horizon logical consistency of LLM-driven interactive narrators, offering a repeatable protocol to compare models on narrative integrity under adversarial user interventions. This supports practical selection of models for interactive storytelling applications."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
