{"record": {"aliases": ["AVCap: Reinforcing Audio-Video Joint Caption with Detail-Aware Reward"], "artifacts": [{"id": "arxiv:2608.06930", "kind": "paper", "url": "https://arxiv.org/abs/2608.06930"}], "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "Evaluates atomic-level details in audiovisual captions using a specialized benchmark and metric, with associated dataset AVCap-100K and model AVCap."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.06930", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.06930", "title": "AVCap-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.06930", "modality": null, "name": "AVCap-Bench", "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_avcap-reinforcing-audio-video-joint-captio_13492b7b", "origin_source": "arxiv", "recognition_confidence": "0.55", "record_sha256": "b07a0df7be030f01de2a7e955485c1a70a3aaa1b6ac3479b772bf4d049975b63", "relation": "evaluates_only", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-05T08:27:33.339760Z", "source_url": "https://arxiv.org/abs/2608.06930"}, "publisher": null, "released": "2026-08-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.06930", "source_url": "https://arxiv.org/abs/2608.06930"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-06930", "source": "claire_radar", "source_benchmark_id": "2608.06930", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["AVCap: Reinforcing Audio-Video Joint Caption with Detail-Aware Reward"], "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": 1, "hfPaperUrl": "https://huggingface.co/papers/2608.06930", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark includes public artifacts on Hugging Face and addresses a gap in detailed audiovisual captioning, likely attracting interest from multimodal research communities.", "score": 65}, "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": "b58a8393eca505097f79c4a26499cabc85678bc672b85aabe1c1227480ad6b17", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "AVCap-Bench and AVCap-Score, a specialized benchmark and metric for evaluating atomic-level details in audiovisual captions", "canonicalNameSource": "abstract", "decisionReason": "The benchmark is explicitly named and described with a metric, and public code, models, and datasets are available via a Hugging Face collection, indicating a practical reuse path.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-05T08:27:33.339760Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates atomic-level details in audiovisual captions using a specialized benchmark and metric, with associated dataset AVCap-100K and model AVCap.", "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", "named entity is a model, method, framework, or dataset used in evaluation"], "snippet": "To address these challenges, we propose: (1) AVCap-100K, a high-quality dataset of 100K temporally aligned, detail-rich audio-video captions; (2) AVCap, a model optimized via Detail-Aware GRPO (Da-GRPO) that achieves state-of-the-art performance among open-source models and matches or surpasses proprietary models on several evaluations; and (3) AVCap-Bench and AVCap-Score, a specialized benchmark and metric for evaluating atomic-level details in audiovisual captions."}, "familyId": "bmf_2e5977a87eb8", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_avcap-reinforcing-audio-video-joint-captio_13492b7b", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.06930", "pdf": "https://arxiv.org/pdf/2608.06930", "project": null, "report": "https://arxiv.org/abs/2608.06930"}, "metrics": [], "motivation": "Detailed audio-video joint captioning is essential for multimodal video understanding and generation.", "name": "AVCap-Bench", "oneLine": "Evaluates atomic-level details in audiovisual captions using a specialized benchmark and metric, with associated dataset AVCap-100K and model AVCap.", "paperTitle": "AVCap: Reinforcing Audio-Video Joint Caption with Detail-Aware Reward", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.06930", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.15, "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": 0.12247537038455211, "value": 1}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 44}, "method": "attention-ranking-v13", "rank": null, "score": 44}}, "readiness": "Paper only", "recognitionConfidence": 0.55, "relation": "evaluates_only", "releasedAt": "2026-08-07", "source": {"authors": ["Mingyang Wu", "Kaituo Feng", "Bohao Li", "Kaixiong Gong", "Zihao Yin", "Xiangyu Yue"], "categories": ["cs.CV"], "id": "2608.06930", "title": "AVCap: Reinforcing Audio-Video Joint Caption with Detail-Aware Reward", "type": "arxiv", "url": "https://arxiv.org/abs/2608.06930"}, "sourceUpdatedAt": "2026-08-07T08:03:51Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Enables fine-grained evaluation of audio-video captioning beyond coarse metrics, addressing a gap for detailed audiovisual understanding and generation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
