{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.06993", "kind": "paper", "url": "https://arxiv.org/abs/2609.06993"}], "authors": ["Sungjune Lee", "Myungjoo Kang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.06993", "source_url": "https://arxiv.org/abs/2609.06993"}, "categories": ["Code & Software", "General AI", "cs.SE"], "description": {"en": "Evaluates Markdown boundary failures, specifically CommonMark-level fence-boundary accuracy, across 4,179 prompts using a scoring CLI that separates content correctness from boundary correctness."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.06993", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.06993", "title": "LatentMD"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Seoul National University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2609.06993", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2609.06993", "source_url": "https://openalex.org/W7212013149"}, "key": "claire-radar:2609.06993", "modality": null, "name": "LatentMD", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_latentmd_cea161f6", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "92b4ab6e2d8361994dc2f73ea7633598449dfa3d5e31ccd0c95a93faae140de1", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-09T06:27:57.312318Z", "source_url": "https://arxiv.org/abs/2609.06993"}, "publisher": null, "released": "2026-09-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.06993", "source_url": "https://arxiv.org/abs/2609.06993"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-06993", "source": "claire_radar", "source_benchmark_id": "2609.06993", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Code & Software", "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": null, "hfPaperUrl": "https://huggingface.co/papers/2609.06993", "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", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "A specialized benchmark targeting Markdown boundary failures in LLMs is likely to attract moderate interest from researchers focused on structured output and format adherence.", "score": 35}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-09T06:27:57.312318Z", "inputHash": "58dd849a19b36ba82756561117fba9367dd0a4cf02137026b1687df2deb8573e", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "We introduce LatentMD, a benchmark and evaluation protocol for diagnosing CommonMark-level fence-boundary failures in LLM-generated Markdown.", "canonicalNameSource": "abstract", "decisionReason": "The paper explicitly releases a benchmark with a defined task, a reproducible scoring protocol via CLI, and a public dataset of 4,179 prompts, meeting criteria for public reuse.", "evaluationEvidence": {"reuse": "The benchmark contains 4,179 prompts and a CLI for scoring arbitrary model outputs.", "scoring": "Separates content correctness from boundary correctness using a CLI for scoring arbitrary model outputs.", "task": "Diagnosing CommonMark-level fence-boundary failures in LLM-generated Markdown."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-09T06:27:57.312318Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates Markdown boundary failures, specifically CommonMark-level fence-boundary accuracy, across 4,179 prompts using a scoring CLI that separates content correctness from boundary correctness.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce LatentMD, a benchmark and evaluation protocol for diagnosing CommonMark-level fence-boundary failures in LLM-generated Markdown."}, "familyId": "bmf_1e1ebbc7e625", "firstSeenAt": "2026-09-09", "heat": null, "id": "bm_latentmd_cea161f6", "indexedAt": "2026-09-09T06:18:29.794093Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.06993", "pdf": "https://arxiv.org/pdf/2609.06993", "project": null, "report": "https://arxiv.org/abs/2609.06993"}, "metrics": [], "motivation": "Large language models (LLMs) increasingly generate Markdown that is consumed by renderers, agents, code extractors, and structured downstream pipelines.", "name": "LatentMD", "oneLine": "Evaluates Markdown boundary failures, specifically CommonMark-level fence-boundary accuracy, across 4,179 prompts using a scoring CLI that separates content correctness from boundary correctness.", "paperTitle": "LatentMD: Benchmarking Markdown Boundary Failures in LLM-Generated Text", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.06993", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-09T06:18:29.794093Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.0, "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": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}, "90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-07", "source": {"authors": ["Sungjune Lee", "Myungjoo Kang"], "categories": ["cs.SE", "cs.CL", "cs.LG"], "id": "2609.06993", "title": "LatentMD: Benchmarking Markdown Boundary Failures in LLM-Generated Text", "type": "arxiv", "url": "https://arxiv.org/abs/2609.06993"}, "sourceUpdatedAt": "2026-09-07T03:35:26Z", "topics": ["cs.SE"], "venueAttempts": [], "whyItMatters": "Existing evaluations conflate content quality with format adherence, missing outputs that are content-correct but boundary-broken; LatentMD isolates this gap for parser-sensitive LLM evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
