{"record": {"aliases": ["Putting Captions to the Test: Evaluating Video Caption Quality through Multiple-Choice Question Answering"], "artifacts": [{"id": "arxiv:2609.09973", "kind": "paper", "url": "https://arxiv.org/abs/2609.09973"}], "authors": ["Zizhen Wang", "Bo Feng", "Zhengfeng Lai", "Shiyu Li", "Yang Lu", "Meng Cao", "Ping Huang", "Xiaoming Simon Wang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.09973", "source_url": "https://arxiv.org/abs/2609.09973"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "Evaluates video caption quality through human-verified multiple-choice questions across 10 types and 24 domains, measuring information fidelity as coverage and factuality."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.09973", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.09973", "title": "CapQuiz"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.09973", "modality": null, "name": "CapQuiz", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "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_putting-captions-to-the-test-evaluating-vi_efc92f10", "origin_source": "arxiv", "recognition_confidence": "0.45", "record_sha256": "17326a7ee787bffb5bd539b65526236a5141113cce2ecbc8b12b1f64274c37bb", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-11T00:52:53.785364Z", "source_url": "https://arxiv.org/abs/2609.09973"}, "publisher": null, "released": "2026-09-09", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.09973", "source_url": "https://arxiv.org/abs/2609.09973"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-09973", "source": "claire_radar", "source_benchmark_id": "2609.09973", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["Putting Captions to the Test: Evaluating Video Caption Quality through Multiple-Choice Question Answering"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-09-10", "attemptedAt": "2026-09-10T12:24:43.080049Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.09973", "observedAt": "2026-09-09T12:31:07.558041Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-10T12:24:43.080049Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "Video caption evaluation is a niche but active area, and the reference-free approach may draw interest from VLLM researchers.", "score": 45}, "capabilities": [], "confidence": "Medium", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-11T00:52:53.785364Z", "inputHash": "8f629c850cf8731aab85f876521476cd0fcfacd1234d6b99dc4b4b0a778222a2", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "We introduce CapQuiz, a novel reference-free benchmark that assesses captions based on their utility in answering human-verified, fine-grained, multiple-choice questions derived from the video.", "canonicalNameSource": "abstract", "decisionReason": "Formally named benchmark with clear task and scoring, though no explicit code/data link; reusable protocol is implied.", "evaluationEvidence": {"reuse": "CapQuiz features a hierarchical taxonomy of 10 question types (spanning Descriptive and Inferential categories) across 24 diverse video domains.", "scoring": "CapQuiz correlates significantly better with human judgments than existing metrics", "task": "assesses captions based on their utility in answering human-verified, fine-grained, multiple-choice questions derived from the video"}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-11T00:52:53.785364Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates video caption quality through human-verified multiple-choice questions across 10 types and 24 domains, measuring information fidelity as coverage and factuality.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence"], "snippet": "We introduce CapQuiz, a novel reference-free benchmark that assesses captions based on their utility in answering human-verified, fine-grained, multiple-choice questions derived from the video."}, "familyId": "bmf_14baa045c3be", "firstSeenAt": "2026-09-10", "heat": null, "id": "bm_putting-captions-to-the-test-evaluating-vi_efc92f10", "indexedAt": "2026-09-10T00:47:26.889471Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.09973", "pdf": "https://arxiv.org/pdf/2609.09973", "project": null, "report": "https://arxiv.org/abs/2609.09973"}, "metrics": [], "motivation": "Evaluating video captioning remains a critical challenge for Visual Large Language Models (VLLMs).", "name": "CapQuiz", "oneLine": "Evaluates video caption quality through human-verified multiple-choice questions across 10 types and 24 domains, measuring information fidelity as coverage and factuality.", "paperTitle": "Putting Captions to the Test: Evaluating Video Caption Quality through Multiple-Choice Question Answering", "primaryDomain": "General AI", "publication": {"evidence": "Accepted by ACL 2026 main conference", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2609.09973", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "ACL 2026 main conference", "verifiedAt": "2026-09-10T00:47:26.889471Z"}, "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": null, "historyDate": null, "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}, "today": {"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": 1, "historyDate": "2026-09-09", "rank": null, "score": null, "windowDays": 1}, "level": {"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}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.56, "role": "bonus", "value": 56}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": 14}, "method": "attention-ranking-v13", "rank": null, "score": 14}}, "readiness": "Paper only", "recognitionConfidence": 0.45, "relation": "introduces", "releasedAt": "2026-09-09", "source": {"authors": ["Zizhen Wang", "Bo Feng", "Zhengfeng Lai", "Shiyu Li", "Yang Lu", "Meng Cao", "Ping Huang", "Xiaoming Simon Wang"], "categories": ["cs.CV"], "id": "2609.09973", "title": "Putting Captions to the Test: Evaluating Video Caption Quality through Multiple-Choice Question Answering", "type": "arxiv", "url": "https://arxiv.org/abs/2609.09973"}, "sourceUpdatedAt": "2026-09-09T10:00:43Z", "topics": ["cs.CV"], "venueAttempts": [{"decisionRaw": "Accepted by ACL 2026 main conference", "evidence": [{"level": "author-claim", "observedAt": "2026-09-10T00:47:26.889471Z", "rawValue": "Accepted by ACL 2026 main conference", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2609.09973"}], "reviewStatus": "accepted", "venueName": "ACL 2026 main conference"}], "whyItMatters": "Offers a reference-free, interpretable alternative to text-matching metrics, enabling finer-grained evaluation of video captioning models."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
