{"record": {"aliases": ["GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks"], "artifacts": [{"id": "arxiv:2608.07411", "kind": "paper", "url": "https://arxiv.org/abs/2608.07411"}, {"id": "gh:rfr2003/geobenchllm", "kind": "repo", "url": "https://github.com/Rfr2003/GeoBenchLLM"}], "authors": ["Rodrigo Ferreira Rodrigues", "Karim Radouane", "Jose G Moreno", "Lynda Tamine"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.07411", "source_url": "https://arxiv.org/abs/2608.07411"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "Evaluates LLMs on 17 geo-related subdatasets across 8 tasks covering knowledge, reasoning, and application, with inference and evaluation scripts."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.07411", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.07411", "title": "GeoBenchLLM"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.07411", "modality": null, "name": "GeoBenchLLM", "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": "true", "origin_record_id": "bm_geobenchllm-a-comprehensive-benchmark-for-_dfb41079", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "5d6cc74348e53eb65c6fe54b6ed319445ca7e3d6ea38eb23954a11dafe66a01b", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-05T08:27:33.339760Z", "source_url": "https://arxiv.org/abs/2608.07411"}, "publisher": null, "released": "2026-08-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.07411", "source_url": "https://arxiv.org/abs/2608.07411"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-07411", "source": "claire_radar", "source_benchmark_id": "2608.07411", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Rfr2003/GeoBenchLLM", "githubScope": "benchmark_repo", "githubStars": 1, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2608.07411", "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"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "A new geo-focused benchmark with public code may attract niche attention from researchers in geo-AI and NLP, though limited by the specific domain.", "score": 35}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-05T08:27:33.339760Z", "inputHash": "ccfbda3cb9987c436f24b101e6d3fb33cee3f672798d2b3ca225ddb6a0aa44a8", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks", "canonicalNameSource": "paper_title", "decisionReason": "Public code repository, released datasets, and evaluation scripts define a repeatable benchmark with clear scoring across multiple geo tasks.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-05T08:27:33.339760Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates LLMs on 17 geo-related subdatasets across 8 tasks covering knowledge, reasoning, and application, with inference and evaluation scripts.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "In this paper, we present \\benchName, a comprehensive benchmark for probing LLMs on geo-related tasks."}, "familyId": "bmf_92fa8f9488b7", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_geobenchllm-a-comprehensive-benchmark-for-_dfb41079", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/Rfr2003/GeoBenchLLM", "data": null, "paper": "https://arxiv.org/abs/2608.07411", "pdf": "https://arxiv.org/pdf/2608.07411", "project": null, "report": "https://arxiv.org/abs/2608.07411"}, "metrics": [], "motivation": "In the context of geodata, existing Large Language Models have often been studied in a homogeneous setting, which has considerably limited insights into their generalization capabilities.", "name": "GeoBenchLLM", "oneLine": "Evaluates LLMs on 17 geo-related subdatasets across 8 tasks covering knowledge, reasoning, and application, with inference and evaluation scripts.", "paperTitle": "GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks", "primaryDomain": "General AI", "publication": {"evidence": "Accepted at CIKM2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2608.07411", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "CIKM2026", "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "publishers": [{"name": "Rodrigo Ferreira Rodrigues", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Rfr2003/GeoBenchLLM"}], "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": 0.09252268850709358, "value": 1}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Medium", "coverage": 0.7, "rank": 556, "score": 20}, "method": "attention-ranking-v13", "rank": 556, "score": 20}}, "readiness": "Runnable", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-08-07", "source": {"authors": ["Rodrigo Ferreira Rodrigues and Karim Radouane and Jose G Moreno and Lynda Tamine"], "categories": ["cs.AI", "cs.CL", "cs.IR", "cs.LG"], "id": "2608.07411", "title": "GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks", "type": "arxiv", "url": "https://arxiv.org/abs/2608.07411"}, "sourceUpdatedAt": "2026-08-07T16:58:04Z", "topics": ["cs.AI"], "venueAttempts": [{"decisionRaw": "Accepted at CIKM2026", "evidence": [{"level": "author-claim", "observedAt": "2026-08-19T11:05:13.395391Z", "rawValue": "Accepted at CIKM2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2608.07411"}], "reviewStatus": "accepted", "venueName": "CIKM2026"}], "whyItMatters": "Provides multi-task geo-spatial evaluation spanning factual, reasoning, and applied capabilities, enabling standardized comparison of LLMs on geo data."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
