{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.22232", "kind": "paper", "url": "https://arxiv.org/abs/2608.22232"}], "authors": ["Zhiming Yang", "Zhuoxi Xiong", "Donglin Zhou", "Wenjun Wei", "Shiyao Cui", "Jinqiao Shi"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.22232", "source_url": "https://arxiv.org/abs/2608.22232"}, "categories": ["General AI", "Multimodal", "Reasoning"], "description": {"en": "Evaluates multimodal large language models on discrimination, understanding, and reasoning under situational illusions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.22232", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.22232", "title": "MSIBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Beijing University of Posts and Telecommunications"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.22232", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.22232", "source_url": "https://openalex.org/W7204196920"}, "key": "claire-radar:2608.22232", "modality": null, "name": "MSIBench", "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": "false", "origin_record_id": "bm_msibench_1f439887", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "269c06d92043fc15cfbd01d1a8ac12396d2aa50e927356231618ca9945f067d7", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-08-27T04:18:56.318436Z", "source_url": "https://arxiv.org/abs/2608.22232"}, "publisher": null, "released": "2026-08-23", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.22232", "source_url": "https://arxiv.org/abs/2608.22232"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-22232", "source": "claire_radar", "source_benchmark_id": "2608.22232", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "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/2608.22232", "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"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "The benchmark targets an emerging robustness issue for MLLMs, but the abstract alone offers limited signal for immediate wide adoption.", "score": 38}, "capabilities": ["Reasoning"], "confidence": "Medium", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-27T04:18:56.318436Z", "inputHash": "0f412802d806ce00f4d1fb4075b8484e6a6c4a057f31fa26a052e9e7138fbf7c", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "we introduce MSIBench, a benchmark designed to assess the discrimination, understanding, and reasoning capabilities of MLLMs under situational illusions", "canonicalNameSource": "abstract", "decisionReason": "No public release evidence beyond the paper; no code, data, or project link is supplied to verify a stable scoring contract or reuse path.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-27T04:18:56.318436Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates multimodal large language models on discrimination, understanding, and reasoning under situational illusions.", "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence"], "snippet": "Building on this taxonomy, we introduce MSIBench, a benchmark designed to assess the discrimination, understanding, and reasoning capabilities of MLLMs under situational illusions."}, "familyId": "bmf_0511d5a6a43f", "firstSeenAt": "2026-08-26", "heat": null, "id": "bm_msibench_1f439887", "indexedAt": "2026-08-26T06:08:22.244942Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.22232", "pdf": "https://arxiv.org/pdf/2608.22232", "project": null, "report": "https://arxiv.org/abs/2608.22232"}, "metrics": [], "motivation": "Real-world situation appearances can deviate from their underlying physical states, challenging the reliability of multimodal large language models (MLLMs) in practical applications.", "name": "MSIBench", "oneLine": "Evaluates multimodal large language models on discrimination, understanding, and reasoning under situational illusions.", "paperTitle": "Beyond What Meets the Eye: Unveiling Situational Illusions for Multimodal Large Language Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.22232", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-26T06:08:22.244942Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-08-23", "source": {"authors": ["Zhiming Yang", "Zhuoxi Xiong", "Donglin Zhou", "Wenjun Wei", "Shiyao Cui", "Jinqiao Shi"], "categories": ["cs.AI", "cs.CL", "cs.CV", "cs.MM"], "id": "2608.22232", "title": "Beyond What Meets the Eye: Unveiling Situational Illusions for Multimodal Large Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2608.22232"}, "sourceUpdatedAt": "2026-08-25T05:51:53Z", "topics": ["Multimodal", "Reasoning"], "venueAttempts": [], "whyItMatters": "Tests whether MLLMs are misled when appearances deviate from physical states, addressing reliability in real-world perception."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
