{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.22725", "kind": "paper", "url": "https://arxiv.org/abs/2608.22725"}], "authors": ["Jiaqi Liu", "Maolin Ran", "Xiaoyang Lu", "Jian Wang", "Weiwen Liu", "Jianghao Lin", "Yong Yu", "Weinan Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.22725", "source_url": "https://arxiv.org/abs/2608.22725"}, "categories": ["General AI", "Vision & 3D", "cs.AI"], "description": {"en": "SEAM-Bench is a double-blind continuity storyboarding benchmark for evaluating visual continuity in short-drama generation. It assesses cross-episode continuity recall and generalizes across six mainstream text models."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.22725", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.22725", "title": "SEAM-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Shanghai Jiao Tong University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.22725", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.22725", "source_url": "https://openalex.org/W7204168363"}, "key": "claire-radar:2608.22725", "modality": null, "name": "SEAM-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_seam-bench_df6f1e82", "origin_source": "arxiv", "recognition_confidence": "0.5", "record_sha256": "9707aa472752e47ac940b49fb2a06361715b974db3cc6ae6f12da3720a26c232", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-08-27T04:18:56.318436Z", "source_url": "https://arxiv.org/abs/2608.22725"}, "publisher": null, "released": "2026-08-24", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.22725", "source_url": "https://arxiv.org/abs/2608.22725"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-22725", "source": "claire_radar", "source_benchmark_id": "2608.22725", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-08-26", "attemptedAt": "2026-08-26T06:23:37.320575Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.22725", "observedAt": "2026-08-24T11:31:30.653116Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-08-26T06:23:37.320575Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "The benchmark is specific to short-drama generation with moderate interest, but its narrow scope may limit initial attention.", "score": 35}, "capabilities": [], "confidence": "Medium", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-27T04:22:25.189955Z", "inputHash": "6b4efb7d483e2a5eca0d8e902962ab8b9392221040573ca9055b12767e27249f", "model": "deepseek-v4-flash"}, "curation": {"canonicalNameEvidence": "We further release SEAM-Bench, a double-blind continuity storyboarding benchmark", "canonicalNameSource": "abstract", "decisionReason": "The benchmark is formally released with a double-blind evaluation setup and is used to compare SEAM against baselines, meeting the criteria for a published benchmark with a stable scoring contract.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-27T04:18:56.318436Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "SEAM-Bench is a double-blind continuity storyboarding benchmark for evaluating visual continuity in short-drama generation. It assesses cross-episode continuity recall and generalizes across six mainstream text models.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract"], "snippet": "We further release SEAM-Bench, a double-blind continuity storyboarding benchmark, on which SEAM raises cross-episode continuity recall from 0.700 to 0.946, generalizes across six mainstream text models, and yields consistent, though not yet significant, gains at the generated-image layer."}, "familyId": "bmf_13bb3b50c414", "firstSeenAt": "2026-08-26", "heat": null, "id": "bm_seam-bench_df6f1e82", "indexedAt": "2026-08-26T06:08:22.244942Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.22725", "pdf": "https://arxiv.org/pdf/2608.22725", "project": null, "report": "https://arxiv.org/abs/2608.22725"}, "metrics": [], "motivation": "Short-drama generation has grown into a large, industrialized pipeline, and as it scales from isolated shots to the episode level, visual continuity has become a critical bottleneck.", "name": "SEAM-Bench", "oneLine": "SEAM-Bench is a double-blind continuity storyboarding benchmark for evaluating visual continuity in short-drama generation. It assesses cross-episode continuity recall and generalizes across six mainstream text models.", "paperTitle": "SEAM: Shot Entity-Attribute Memory for Consistent Short-Drama Generation at Scale", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.22725", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-26T06:08:22.244942Z"}, "publications": [], "publishers": [{"name": "CreativeFitting", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.22725"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"percentile": null, "value": null}, "hfDatasetDownloads": {"percentile": null, "value": null}, "hfPaperUpvotes": {"percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"percentile": null, "value": null}, "hfDatasetDownloads": {"percentile": null, "value": null}, "hfPaperUpvotes": {"percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v12", "rank": null, "score": null}, "90d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"percentile": null, "value": null}, "hfDatasetDownloads": {"percentile": null, "value": null}, "hfPaperUpvotes": {"percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"percentile": null, "value": null}, "hfDatasetDownloads": {"percentile": null, "value": null}, "hfPaperUpvotes": {"percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v12", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 0.5, "relation": "introduces", "releasedAt": "2026-08-24", "source": {"authors": ["Jiaqi Liu", "Maolin Ran", "Xiaoyang Lu", "Jian Wang", "Weiwen Liu", "Jianghao Lin", "Yong Yu", "Weinan Zhang"], "categories": ["cs.AI"], "id": "2608.22725", "title": "SEAM: Shot Entity-Attribute Memory for Consistent Short-Drama Generation at Scale", "type": "arxiv", "url": "https://arxiv.org/abs/2608.22725"}, "sourceUpdatedAt": "2026-08-24T02:27:50Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "The benchmark fills the gap in evaluating visual continuity in large-scale short-drama generation, which is critical for production pipelines. It provides a standardized protocol to compare memory-based approaches and guide improvements in episodic generation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
