{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.03597", "kind": "paper", "url": "https://arxiv.org/abs/2609.03597"}, {"id": "hf:raiyankhaan/khatiandoc", "kind": "dataset", "url": "https://huggingface.co/datasets/RaiyanKhaan/KhatianDoc"}], "authors": ["Tasmiad Hasan", "Arafat Zaman Ratul", "Sarker Sadman Saalim", "S. M. Shah Nawaz Hossain", "Khan Raiyan Ibne Reza", "Sumaiya Tabassum Nimi"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.03597", "source_url": "https://arxiv.org/abs/2609.03597"}, "categories": ["Multimodal", "Software & AI Compute"], "description": {"en": "A four-task benchmark built from 107 real RS Khatian land records evaluating symbol recognition, base-16 conversion, structured field extraction, and legal QA with 1,634 QA pairs."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.03597", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.03597", "title": "KhatianDoc"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["North South University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2609.03597", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2609.03597", "source_url": "https://openalex.org/W7208771419"}, "key": "claire-radar:2609.03597", "modality": null, "name": "KhatianDoc", "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": "true", "origin_record_id": "bm_khatiandoc_7e28ecbb", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "00d74d8d48d4247a3d60e6070729aa89a0a3da62eccb56f7aee8e111da6fc0d9", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-07T00:42:15.394147Z", "source_url": "https://arxiv.org/abs/2609.03597"}, "publisher": null, "released": "2026-09-03", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.03597", "source_url": "https://arxiv.org/abs/2609.03597"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-03597", "source": "claire_radar", "source_benchmark_id": "2609.03597", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Multimodal", "attention": {"asOf": "2026-09-04", "attemptedAt": "2026-09-04T06:36:46.783164Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "RaiyanKhaan/KhatianDoc", "hfDatasetDownloads": 24, "hfDatasetLikes": 0, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.03597", "observedAt": "2026-09-04T06:36:46.783164Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"observedAt": "2026-09-04T06:36:46.783164Z", "state": "fresh"}, "hfDatasetLikes": {"observedAt": "2026-09-04T06:36:46.783164Z", "state": "fresh"}, "hfPaperUpvotes": {"checkedAt": "2026-09-04T06:36:46.783164Z", "state": "unavailable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses a specialized low-resource legal and multimodal challenge with immediate dataset availability, likely drawing moderate attention from multilingual and domain-specific evaluation communities.", "score": 45}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-07T00:42:15.394147Z", "inputHash": "fb3c72eb4ea51d20ba31fccd9a747163150c2e3cc0ff5df63b7710cd502e5b84", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "KhatianDoc: A Human-Verified Benchmark Diagnosing Multimodal LLM Failure on Bengali Legal Land Records", "canonicalNameSource": "paper_title", "decisionReason": "The paper provides a dataset on Hugging Face, describes a fixed evaluation protocol over six models, and includes verified ground truth, supporting a reusable public benchmark.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-07T00:42:15.394147Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A four-task benchmark built from 107 real RS Khatian land records evaluating symbol recognition, base-16 conversion, structured field extraction, and legal QA with 1,634 QA pairs.", "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", "public artifact URL"], "snippet": "We introduce KhatianDoc, a four-task benchmark built from 107 real RS Khatian records from the Vumi (land) Office of Munshiganj, Bangladesh: symbol recognition, base-16-to-decimal conversion, structured field extraction, and legal document question answering over 1,634 QA pairs."}, "familyId": "bmf_7253ab00f3bd", "firstSeenAt": "2026-09-04", "heat": null, "id": "bm_khatiandoc_7e28ecbb", "indexedAt": "2026-09-04T00:41:56.430646Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": "https://huggingface.co/datasets/RaiyanKhaan/KhatianDoc", "paper": "https://arxiv.org/abs/2609.03597", "pdf": "https://arxiv.org/pdf/2609.03597", "project": null, "report": "https://arxiv.org/abs/2609.03597"}, "metrics": [], "motivation": "Land ownership in Bangladesh is recorded in Ana-Ganda-Kora-Kranti-Til, a base-16 positional fraction system with dedicated Unicode glyphs, no mainstream font, and no coverage in any OCR pipeline or tokenizer.", "name": "KhatianDoc", "oneLine": "A four-task benchmark built from 107 real RS Khatian land records evaluating symbol recognition, base-16 conversion, structured field extraction, and legal QA with 1,634 QA pairs.", "paperTitle": "KhatianDoc: A Human-Verified Benchmark Diagnosing Multimodal LLM Failure on Bengali Legal Land Records", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.03597", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-04T00:41:56.430646Z"}, "publications": [], "publishers": [{"name": "RaiyanKhaan", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://huggingface.co/datasets/RaiyanKhaan/KhatianDoc"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.15, "datasetDownloadRank": 36, "datasetRankPopulation": 37, "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": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": 0.3359521503221738, "value": 24}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.15, "rank": 41, "score": 48}, "method": "attention-ranking-v13", "rank": 41, "score": 48}, "90d": {"confidence": "Low", "coverage": 0.3, "datasetDownloadRank": 71, "datasetRankPopulation": 73, "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": 0.26662798503060087, "value": 24}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.3, "rank": 125, "score": 43}, "method": "attention-ranking-v13", "rank": 125, "score": 43}, "today": {"confidence": "Low", "coverage": 0.05, "datasetDownloadRank": 1, "datasetRankPopulation": 1, "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": 1, "historyDate": "2026-09-03", "rank": null, "score": null, "windowDays": 1}, "level": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": 0.5, "percentile": 0.5, "value": 24}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.55, "role": "bonus", "value": 55}}, "confidence": "Low", "coverage": 0.05, "rank": 6, "score": 51}, "method": "attention-ranking-v13", "rank": 6, "score": 51}}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-03", "source": {"authors": ["Tasmiad Hasan", "Arafat Zaman Ratul", "Sarker Sadman Saalim", "S.M. Shah Nawaz Hossain", "Khan Raiyan Ibne Reza and Sumaiya Tabassum Nimi"], "categories": ["cs.CL"], "id": "2609.03597", "title": "KhatianDoc: A Human-Verified Benchmark Diagnosing Multimodal LLM Failure on Bengali Legal Land Records", "type": "arxiv", "url": "https://arxiv.org/abs/2609.03597"}, "sourceUpdatedAt": "2026-09-03T09:46:41Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "It documents an absence of capability in reading Bengali legal land records with a rare base-16 fraction system, offering verified ground truth to guide future model development."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
