{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.09746", "kind": "paper", "url": "https://arxiv.org/abs/2606.09746"}], "authors": ["Sherwin Varghese", "Matthew Wicker", "Alessio Lomuscio"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.09746", "source_url": "https://arxiv.org/abs/2606.09746"}, "categories": ["Autonomous Driving", "Robustness", "Safety & Trustworthiness", "cs.CV"], "description": {"en": "ST-Bench is a verification benchmark for evaluating certified robustness of spatio-temporal neural networks on autonomous driving (Udacity) and activity recognition (UCF-101) tasks, using spatio-temporal perturbation constraints."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.09746", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.09746", "title": "ST-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Imperial College London"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.09746", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.09746", "source_url": "https://openalex.org/W7164025265"}, "key": "claire-radar:2606.09746", "modality": null, "name": "ST-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": "unknown", "origin_record_id": "bm_st-bench_eeee9e2f", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "19b3ac4ad1727fc54e8e9b752daa9ba2ec74d017ddef3be44d04e67f2ca012a0", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.09746"}, "publisher": null, "released": "2026-06-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.09746", "source_url": "https://arxiv.org/abs/2606.09746"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-09746", "source": "claire_radar", "source_benchmark_id": "2606.09746", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Autonomous Driving"], "area": "Safety & Trustworthiness", "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": null, "hfPaperUrl": "https://huggingface.co/papers/2606.09746", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": ["Robustness"], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "85a0aa69d53bb7351576bac24f6aca41b9443a1ff5d32b6f96ce3155c4c63d0c", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ST-Bench is a verification benchmark for evaluating certified robustness of spatio-temporal neural networks on autonomous driving (Udacity) and activity recognition (UCF-101) tasks, using spatio-temporal perturbation constraints.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence"], "snippet": "To spur further progress in this field, we propose ST-Bench, a verification benchmark for autonomous driving and activity recognition, to systematically evaluate verifiable robustness."}, "familyId": "bmf_c12a5ba8d0fc", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_st-bench_eeee9e2f", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Automotive"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.09746", "paper": "https://arxiv.org/abs/2606.09746", "pdf": "https://arxiv.org/pdf/2606.09746", "project": null, "report": "https://arxiv.org/abs/2606.09746"}, "metrics": [], "motivation": "With AI increasingly deployed in safety-critical systems, providing formal robustness guarantees for the underlying models is essential.", "name": "ST-Bench", "oneLine": "ST-Bench is a verification benchmark for evaluating certified robustness of spatio-temporal neural networks on autonomous driving (Udacity) and activity recognition (UCF-101) tasks, using spatio-temporal perturbation constraints.", "paperTitle": "Hybrid Robustness Verification for Spatio-Temporal Neural Networks", "primaryDomain": "Autonomous Driving", "publication": {"evidence": "Accepted at the 9th International Symposium on AI Verification (SAIV 2026)", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.09746", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "9th International Symposium on AI Verification (SAIV 2026)", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-06-08", "source": {"authors": ["Sherwin Varghese", "Matthew Wicker and Alessio Lomuscio"], "categories": ["cs.CV", "cs.AI", "cs.LG"], "id": "2606.09746", "title": "Hybrid Robustness Verification for Spatio-Temporal Neural Networks", "type": "arxiv", "url": "https://arxiv.org/abs/2606.09746"}, "sourceUpdatedAt": "2026-06-08T17:06:51Z", "topics": ["cs.CV"], "venueAttempts": [{"decisionRaw": "Accepted at the 9th International Symposium on AI Verification (SAIV 2026)", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at the 9th International Symposium on AI Verification (SAIV 2026)", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.09746"}], "reviewStatus": "accepted", "venueName": "9th International Symposium on AI Verification (SAIV 2026)"}], "whyItMatters": "Existing robustness verification methods rely on overly conservative assumptions or are computationally prohibitive. ST-Bench provides a realistic, constrained perturbation model for video inputs, enabling tighter approximations and more meaningful robustness comparisons for safety-critical applications."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
