{"record": {"aliases": ["From Visual Cues to Spoken Narration: Rethinking Audio Description"], "artifacts": [{"id": "arxiv:2609.01725", "kind": "paper", "url": "https://arxiv.org/abs/2609.01725"}, {"id": "gh:multimodal-ai-lab/cue2narrate", "kind": "repo", "url": "https://github.com/multimodal-ai-lab/Cue2Narrate"}], "authors": ["Akshita Gupta", "Aditya Arora", "Federico Tombari", "Marcus Rohrbach", "Anna Rohrbach"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.01725", "source_url": "https://arxiv.org/abs/2609.01725"}, "categories": ["General AI", "Speech & Audio", "cs.CV"], "description": {"en": "Evaluates audio description generation on long untrimmed movie clips, covering multi-segment localization and narration quality."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.01725", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.01725", "title": "LongLSMDC"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.01725", "modality": null, "name": "LongLSMDC", "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_from-visual-cues-to-spoken-narration-rethi_b0d7498a", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "5509b7a8df17f7fc976149f455c88518165ba883dede72b8c87b10f2fdb33d83", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-03T06:06:59.591745Z", "source_url": "https://arxiv.org/abs/2609.01725"}, "publisher": null, "released": "2026-09-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.01725", "source_url": "https://arxiv.org/abs/2609.01725"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-01725", "source": "claire_radar", "source_benchmark_id": "2609.01725", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["From Visual Cues to Spoken Narration: Rethinking Audio Description"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Speech & Audio", "attention": {"asOf": "2026-09-03", "attemptedAt": "2026-09-03T01:40:48.194060Z", "githubRepo": "https://github.com/multimodal-ai-lab/Cue2Narrate", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.01725", "observedAt": "2026-09-03T01:40:48.194060Z", "signalStatus": {"githubStars": {"observedAt": "2026-09-03T01:40:48.194060Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-03T01:40:48.194060Z", "state": "unavailable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Addresses a novel long-form video grounding task for accessibility with clear public artifacts and links.", "score": 70}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-03T06:06:59.591745Z", "inputHash": "511eca30712415626807b47ab60dec42071df71a1b6d6236f06f200d40e710a4", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "we introduce the LongLSMDC benchmark", "canonicalNameSource": "abstract", "decisionReason": "Formally introduced benchmark with defined evaluation protocol and public code/data link.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-03T06:06:59.591745Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates audio description generation on long untrimmed movie clips, covering multi-segment localization and narration quality.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To benchmark this new problem statement, we introduce the LongLSMDC benchmark with up to 8-min movie clips (~6.5min on average)."}, "familyId": "bmf_2259106e5586", "firstSeenAt": "2026-09-03", "heat": null, "id": "bm_from-visual-cues-to-spoken-narration-rethi_b0d7498a", "indexedAt": "2026-09-03T00:55:28.123857Z", "industrySectors": [], "links": {"code": "https://github.com/multimodal-ai-lab/Cue2Narrate", "data": null, "paper": "https://arxiv.org/abs/2609.01725", "pdf": "https://arxiv.org/pdf/2609.01725", "project": null, "report": "https://arxiv.org/abs/2609.01725"}, "metrics": [], "motivation": "Audio Description (AD) provides spoken narration of visual events during dialogue gaps, making movies accessible to visually impaired audiences.", "name": "LongLSMDC", "oneLine": "Evaluates audio description generation on long untrimmed movie clips, covering multi-segment localization and narration quality.", "paperTitle": "From Visual Cues to Spoken Narration: Rethinking Audio Description", "primaryDomain": "General AI", "publication": {"evidence": "Accepted to EMNLP main conference", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2609.01725", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "EMNLP main conference", "verifiedAt": "2026-09-03T00:55:28.123857Z"}, "publications": [], "publishers": [{"name": "Multimodal AI Lab", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/multimodal-ai-lab/Cue2Narrate"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.55, "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": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 126, "score": 23}, "method": "attention-ranking-v13", "rank": 126, "score": 23}, "90d": {"confidence": "Low", "coverage": 0.55, "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": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 317, "score": 23}, "method": "attention-ranking-v13", "rank": 317, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-09-01", "source": {"authors": ["Akshita Gupta", "Aditya Arora", "Federico Tombari", "Marcus Rohrbach", "Anna Rohrbach"], "categories": ["cs.CV"], "id": "2609.01725", "title": "From Visual Cues to Spoken Narration: Rethinking Audio Description", "type": "arxiv", "url": "https://arxiv.org/abs/2609.01725"}, "sourceUpdatedAt": "2026-09-01T18:00:50Z", "topics": ["cs.CV"], "venueAttempts": [{"decisionRaw": "Accepted to EMNLP main conference", "evidence": [{"level": "author-claim", "observedAt": "2026-09-03T00:55:28.123857Z", "rawValue": "Accepted to EMNLP main conference", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2609.01725"}], "reviewStatus": "accepted", "venueName": "EMNLP main conference"}], "whyItMatters": "Extends audio description beyond pre-segmented clips, requiring joint prediction of visual cue and narration timing in long-form video."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
