{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.05703", "kind": "paper", "url": "https://arxiv.org/abs/2608.05703"}, {"id": "gh:jia-lab-research/streamarena", "kind": "repo", "url": "https://github.com/JIA-Lab-research/StreamArena"}], "authors": ["Xichen Zhang", "Guankai Li", "Yinghao Zhu", "Shijian Wang", "Sitong Wu", "Shaozuo Yu", "Meng Chu", "Yuan Lu", "Jiaya Jia"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.05703", "source_url": "https://arxiv.org/abs/2608.05703"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "StreamArena evaluates hour-scale streaming video understanding across 243 full-length videos (avg 88.8 min) with 3,646 open-ended QA pairs, covering real-time perception, historical retrospection, proactive interaction, and multimodal tool use. Includes a standardized runner and LLM-as-judge scorer."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.05703", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.05703", "title": "StreamArena"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Hong Kong University of Science and Technology", "Xiaohongshu (Little Red Book) (China)", "University of Hong Kong", "Chinese University of Hong Kong"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.05703", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.05703", "source_url": "https://openalex.org/W7201864904"}, "key": "claire-radar:2608.05703", "modality": null, "name": "StreamArena", "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_streamarena_4967785a", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "adf2a9583c41c178491287efa4e91a0140038522e32f30d20bae0e93b67b0ed5", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.05703"}, "publisher": null, "released": "2026-08-06", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.05703", "source_url": "https://arxiv.org/abs/2608.05703"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-05703", "source": "claire_radar", "source_benchmark_id": "2608.05703", "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": "https://github.com/JIA-Lab-research/StreamArena", "githubScope": "benchmark_repo", "githubStars": 34, "hfDailySubmittedAt": "2026-08-10T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 17, "hfPaperUrl": "https://huggingface.co/papers/2608.05703", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "671f0a7dc9f6685e9fefd6ba2ee599a227ef4daf4d22af3efd839b23f53bcd68", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "StreamArena evaluates hour-scale streaming video understanding across 243 full-length videos (avg 88.8 min) with 3,646 open-ended QA pairs, covering real-time perception, historical retrospection, proactive interaction, and multimodal tool use. Includes a standardized runner and LLM-as-judge scorer.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence"], "snippet": "We introduce StreamArena, a benchmark for hour-scale, interactive streaming video understanding."}, "familyId": "bmf_5aa1f09351b6", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/JIA-Lab-research/StreamArena", "url": "https://github.com/JIA-Lab-research/StreamArena"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/JIA-Lab-research/StreamArena", "url": "https://huggingface.co/papers/2608.05703"}], "scope": "benchmark_repo", "url": "https://github.com/JIA-Lab-research/StreamArena", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_streamarena_4967785a", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/JIA-Lab-research/StreamArena", "data": null, "hfPaper": "https://huggingface.co/papers/2608.05703", "paper": "https://arxiv.org/abs/2608.05703", "pdf": "https://arxiv.org/pdf/2608.05703", "project": null, "report": "https://arxiv.org/abs/2608.05703"}, "metrics": [], "motivation": "Deploying autonomous multimodal agents in continuous, real-world environments requires them to ingest unbounded audio-visual streams and maintain hour-scale memory.", "name": "StreamArena", "oneLine": "StreamArena evaluates hour-scale streaming video understanding across 243 full-length videos (avg 88.8 min) with 3,646 open-ended QA pairs, covering real-time perception, historical retrospection, proactive interaction, and multimodal tool use. Includes a standardized runner and LLM-as-judge scorer.", "paperTitle": "StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.05703", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "JIA-Lab-research", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/JIA-Lab-research/StreamArena"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "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": 0.47457505484152435, "value": 34}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.5107131090274512, "value": 17}}, "confidence": "Medium", "coverage": 0.7, "rank": 73, "score": 49}, "method": "attention-ranking-v13", "rank": 73, "score": 49}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-06", "source": {"authors": ["Xichen Zhang", "Guankai Li", "Yinghao Zhu", "Shijian Wang", "Sitong Wu", "Shaozuo Yu", "Meng Chu", "Yuan Lu", "Jiaya Jia"], "categories": ["cs.CV"], "id": "2608.05703", "title": "StreamArena: Toward Continuous, Interactive, and Long-Horizon Agentic Streaming Video Understanding", "type": "arxiv", "url": "https://arxiv.org/abs/2608.05703"}, "sourceUpdatedAt": "2026-08-06T07:46:37Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Addresses the lack of benchmarks for long-horizon, interactive streaming video understanding, where short clips and multiple-choice formats allow shortcuts. Provides a rigorous, open-ended evaluation to assess progress in continuous, interactive multimodal agents."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
