{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.15768", "kind": "paper", "url": "https://arxiv.org/abs/2607.15768"}, {"id": "gh:intellisensing/geochrono", "kind": "repo", "url": "https://github.com/IntelliSensing/GeoChrono"}], "authors": ["Yujie Li", "Jiancheng Pan", "Zhiwei Wei", "Jiuniu Wang", "Mugen Peng", "Wenjia Xu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.15768", "source_url": "https://arxiv.org/abs/2607.15768"}, "categories": ["General AI", "Reasoning", "Vision & 3D"], "description": {"en": "ChronoBench evaluates long-term temporal understanding in remote sensing across four progressive cognitive levels: land cover perception, temporal recognition, long-term memory, and spatio-temporal reasoning. It comprises 12 sub-tasks and 17,689 QA pairs over 3,469 images spanning 500 regions across 39 U.S. cities."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.15768", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.15768", "title": "ChronoBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.15768", "modality": null, "name": "ChronoBench", "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_chronobench_44b7b76d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "0c4b06efe8f9fa0422fc1fcf8aefa99b413947806ebecc0059bdd1f73a695f48", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.15768"}, "publisher": null, "released": "2026-07-17", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.15768", "source_url": "https://arxiv.org/abs/2607.15768"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-15768", "source": "claire_radar", "source_benchmark_id": "2607.15768", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/IntelliSensing/GeoChrono", "githubScope": "benchmark_repo", "githubStars": 11, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2607.15768", "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": ["Reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "4209e15b06135587f7b1fe72cd1383fc1680fe72216c0a023a38d9dcacd2c4db", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ChronoBench evaluates long-term temporal understanding in remote sensing across four progressive cognitive levels: land cover perception, temporal recognition, long-term memory, and spatio-temporal reasoning. It comprises 12 sub-tasks and 17,689 QA pairs over 3,469 images spanning 500 regions across 39 U.S. cities.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To fill this gap, we introduce ChronoBench, a multidimensional benchmark that decomposes this task into four progressive cognitive levels (i.e., Land Cover Perception, Temporal Recognition, Long-Term Memory, and Spatio-Temporal Reasoning)."}, "familyId": "bmf_384c2b00f08c", "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/IntelliSensing/GeoChrono", "url": "https://github.com/IntelliSensing/GeoChrono"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/IntelliSensing/GeoChrono", "url": "https://huggingface.co/papers/2607.15768"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/IntelliSensing/GeoChrono", "url": "https://huggingface.co/api/papers/2607.15768"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/IntelliSensing/GeoChrono", "url": "https://arxiv.org/abs/2607.15768"}], "scope": "project_repo", "url": "https://github.com/IntelliSensing/GeoChrono", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_chronobench_44b7b76d", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/IntelliSensing/GeoChrono", "data": null, "hfPaper": "https://huggingface.co/papers/2607.15768", "paper": "https://arxiv.org/abs/2607.15768", "pdf": "https://arxiv.org/pdf/2607.15768", "project": null, "report": "https://arxiv.org/abs/2607.15768"}, "metrics": [], "motivation": "Remote sensing offers an unparalleled vantage point for observing the Earth's long-term surface evolution, yet it demands that a model not only perceive land cover at isolated moments, but also track changes, memorize evolution histories, and reason across time and space.", "name": "ChronoBench", "oneLine": "ChronoBench evaluates long-term temporal understanding in remote sensing across four progressive cognitive levels: land cover perception, temporal recognition, long-term memory, and spatio-temporal reasoning. It comprises 12 sub-tasks and 17,689 QA pairs over 3,469 images spanning 500 regions across 39 U.S. cities.", "paperTitle": "GeoChrono: Benchmarking and Rethinking Long-Term Temporal Understanding in Remote Sensing", "primaryDomain": "General AI", "publication": {"evidence": "Accepted to ACM MM 2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2607.15768", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "ACM MM 2026", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "IntelliSensing Lab", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/IntelliSensing/GeoChrono"}], "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": 0.3316903687638348, "value": 11}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.12247537038455211, "value": 1}}, "confidence": "Medium", "coverage": 0.7, "rank": 187, "score": 35}, "method": "attention-ranking-v13", "rank": 187, "score": 35}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-17", "source": {"authors": ["Yujie Li", "Jiancheng Pan", "Zhiwei Wei", "Jiuniu Wang", "Mugen Peng", "Wenjia Xu"], "categories": ["cs.CV", "cs.AI"], "id": "2607.15768", "title": "GeoChrono: Benchmarking and Rethinking Long-Term Temporal Understanding in Remote Sensing", "type": "arxiv", "url": "https://arxiv.org/abs/2607.15768"}, "sourceUpdatedAt": "2026-07-17T08:58:29Z", "topics": ["Reasoning"], "venueAttempts": [{"decisionRaw": "Accepted to ACM MM 2026", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted to ACM MM 2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2607.15768"}], "reviewStatus": "accepted", "venueName": "ACM MM 2026"}], "whyItMatters": "Existing remote sensing benchmarks typically focus on static or bi-temporal analysis, lacking a systematic dissection of long-term temporal competencies. ChronoBench provides a multidimensional evaluation that isolates specific cognitive bottlenecks, enabling targeted model improvement. Its integration with lmms-eval supports standardized comparison of multimodal LLMs for satellite image time series."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
