{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.04397", "kind": "paper", "url": "https://arxiv.org/abs/2608.04397"}, {"id": "gh:hae-rae/nolli", "kind": "repo", "url": "https://github.com/HAE-RAE/NOLLI"}], "authors": ["Dasol Choi", "Joonyong Park", "Daegon Yu", "Soo Yong Kim", "Youngsook Song", "Seunghyeok Hong"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.04397", "source_url": "https://arxiv.org/abs/2608.04397"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "NOLLI is a procedurally generated English-Korean puzzle benchmark with 15 puzzle types (25 tasks, 7,500 items). Each instance is seed-regenerable, verified to have a unique solution, and scored deterministically. Difficulty is calibrated behaviorally to target accuracy bands."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.04397", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.04397", "title": "NOLLI"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Korea Telecom (South Korea)", "Hankuk University of Foreign Studies"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.04397", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.04397", "source_url": "https://openalex.org/W7196930622"}, "key": "claire-radar:2608.04397", "modality": null, "name": "NOLLI", "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_nolli_b5bb5859", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "e973083b0f804a6982915e83c751c8ec3627cc759005c198877f113403922602", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.04397"}, "publisher": null, "released": "2026-08-05", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.04397", "source_url": "https://arxiv.org/abs/2608.04397"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-04397", "source": "claire_radar", "source_benchmark_id": "2608.04397", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Machine Generated", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/HAE-RAE/NOLLI", "githubScope": "benchmark_repo", "githubStars": 5, "hfDailySubmittedAt": "2026-08-06T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 22, "hfPaperUrl": "https://huggingface.co/papers/2608.04397", "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": "High", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "d5b263a38d9fcdd605bc1a4663b7748e42e1fd043a37ffb89972162a092a1fd3", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "NOLLI is a procedurally generated English-Korean puzzle benchmark with 15 puzzle types (25 tasks, 7,500 items). Each instance is seed-regenerable, verified to have a unique solution, and scored deterministically. Difficulty is calibrated behaviorally to target accuracy bands.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce NOLLI, a procedurally generated English-Korean puzzle benchmark designed to diagnose where Korean performance gaps arise."}, "familyId": "bmf_9e92e28bd0ca", "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/HAE-RAE/NOLLI", "url": "https://github.com/HAE-RAE/NOLLI"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/HAE-RAE/NOLLI", "url": "https://huggingface.co/papers/2608.04397"}], "scope": "benchmark_repo", "url": "https://github.com/HAE-RAE/NOLLI", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_nolli_b5bb5859", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/HAE-RAE/NOLLI", "data": null, "hfPaper": "https://huggingface.co/papers/2608.04397", "paper": "https://arxiv.org/abs/2608.04397", "pdf": "https://arxiv.org/pdf/2608.04397", "project": null, "report": "https://arxiv.org/abs/2608.04397"}, "metrics": [], "motivation": "We introduce NOLLI, a procedurally generated English-Korean puzzle benchmark designed to diagnose where Korean performance gaps arise.", "name": "NOLLI", "oneLine": "NOLLI is a procedurally generated English-Korean puzzle benchmark with 15 puzzle types (25 tasks, 7,500 items). Each instance is seed-regenerable, verified to have a unique solution, and scored deterministically. Difficulty is calibrated behaviorally to target accuracy bands.", "paperTitle": "NOLLI: A Difficulty-Calibrated Puzzle Benchmark for Diagnosing the English-Korean Performance Gap", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.04397", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "HAE-RAE", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/HAE-RAE/NOLLI"}], "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": 0.2391676802567412, "value": 5}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.5540249260255516, "value": 22}}, "confidence": "Medium", "coverage": 0.7, "rank": 167, "score": 36}, "method": "attention-ranking-v13", "rank": 167, "score": 36}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-05", "source": {"authors": ["Dasol Choi", "Joonyong Park", "Daegon Yu", "Soo Yong Kim", "Youngsook Song", "Seunghyeok Hong"], "categories": ["cs.CL"], "id": "2608.04397", "title": "NOLLI: A Difficulty-Calibrated Puzzle Benchmark for Diagnosing the English-Korean Performance Gap", "type": "arxiv", "url": "https://arxiv.org/abs/2608.04397"}, "sourceUpdatedAt": "2026-08-05T02:58:53Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "NOLLI addresses the lack of controlled cross-lingual benchmarks that separate presentation language from reasoning difficulty, enabling diagnosis of where model performance gaps arise across languages and writing systems."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
