{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.25445", "kind": "paper", "url": "https://arxiv.org/abs/2606.25445"}], "authors": ["Jae-Woo Kim", "Hyeongbeom Kim", "Ue-Hwan Kim"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.25445", "source_url": "https://arxiv.org/abs/2606.25445"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "C3-Bench evaluates context-aware change captioning with 4,996 human-labeled image pairs across 51 real-world contexts, using an LLM-as-Judge framework scoring correctness, specificity, fluency, relevance, and a reversibility metric."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.25445", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.25445", "title": "C3-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.25445", "modality": null, "name": "C3-Bench", "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_c3-bench_c82d4fcd", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "858d5d7655fc73bc021e949ffd237790477da4f8b4505a1fa72827d3c618b0c5", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.25445"}, "publisher": null, "released": "2026-06-24", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.25445", "source_url": "https://arxiv.org/abs/2606.25445"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-25445", "source": "claire_radar", "source_benchmark_id": "2606.25445", "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": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.25445", "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"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "30ec2ca6804caee6ea6be21a47bec4b8e687d6de1e05519e737c92956db04115", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "C3-Bench evaluates context-aware change captioning with 4,996 human-labeled image pairs across 51 real-world contexts, using an LLM-as-Judge framework scoring correctness, specificity, fluency, relevance, and a reversibility metric.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To fill this gap, we propose C3-Bench, a comprehensive benchmark for evaluating Context-aware Change Captioning."}, "familyId": "bmf_8434d033ab49", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_c3-bench_c82d4fcd", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.25445", "paper": "https://arxiv.org/abs/2606.25445", "pdf": "https://arxiv.org/pdf/2606.25445", "project": null, "report": "https://arxiv.org/abs/2606.25445"}, "metrics": [], "motivation": "While Change Captioning systems have garnered substantial attention to respond to our evolving world, their true performance on diverse real-world change contexts remains largely unexplored due to the lack of comprehensive evaluation frameworks.", "name": "C3-Bench", "oneLine": "C3-Bench evaluates context-aware change captioning with 4,996 human-labeled image pairs across 51 real-world contexts, using an LLM-as-Judge framework scoring correctness, specificity, fluency, relevance, and a reversibility metric.", "paperTitle": "C3-Bench: A Context-Aware Change Captioning Benchmark", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.25445", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "arXiv", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2606.25445"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-24", "source": {"authors": ["Jae-Woo Kim", "Hyeongbeom Kim", "Ue-Hwan Kim"], "categories": ["cs.CV", "cs.AI"], "id": "2606.25445", "title": "C3-Bench: A Context-Aware Change Captioning Benchmark", "type": "arxiv", "url": "https://arxiv.org/abs/2606.25445"}, "sourceUpdatedAt": "2026-06-24T06:15:24Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Change captioning performance varies with domain; C3-Bench exposes systematic failures in conventional models and LMMs, providing a comprehensive benchmark to drive generalization and trustworthiness."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
