{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.12606", "kind": "paper", "url": "https://arxiv.org/abs/2609.12606"}], "authors": ["Zhitong Dong", "Jicai Pan", "Yingguo Gao", "Jingting Ding", "Hao Chen", "Jinjie Gu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.12606", "source_url": "https://arxiv.org/abs/2609.12606"}, "categories": ["Geometric reasoning", "Language & Knowledge", "Reasoning", "Robotics & Embodied AI"], "description": {"en": "GeoVAD-Bench diagnoses visual chain-of-thought for geometry problem solving along five dimensions: perception, auxiliary quality, utilization, deductive reasoning, and final correctness, with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.12606", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.12606", "title": "GeoVAD-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Southeast University", "Ant Group (China)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2609.12606", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2609.12606", "source_url": "https://openalex.org/W7212504398"}, "key": "claire-radar:2609.12606", "modality": null, "name": "GeoVAD-Bench", "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": "false", "origin_record_id": "bm_geovad-bench_dadd2d8d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "dfc4e23d4c8dbb97a0dbb9c548c05c4d0916f2a6f12a7790869528a0bffe316f", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-14T00:43:53.625753Z", "source_url": "https://arxiv.org/abs/2609.12606"}, "publisher": null, "released": "2026-09-11", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.12606", "source_url": "https://arxiv.org/abs/2609.12606"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-12606", "source": "claire_radar", "source_benchmark_id": "2609.12606", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Robotics & Embodied AI"], "area": "Language & Knowledge", "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": 7, "hfPaperUrl": "https://huggingface.co/papers/2609.12606", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "A named diagnostic geometry benchmark with a five-dimensional evaluation scheme and intervention settings could draw multimodal reasoning interest, though no artifacts or public path are supplied.", "score": 22}, "capabilities": ["Reasoning", "Geometric reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-15T09:37:10.086794Z", "inputHash": "54cbcf5b789eb594fd6289203cde170b42d23f59ab32af49ae0eec4902ad8d31", "model": "deepseek-v4-flash"}, "curation": {"decisionReason": "The paper introduces the benchmark but no artifact or public path is provided; without released data or code the benchmark cannot be reused.", "model": "deepseek-v4-pro", "policy": "source-evidence-v2", "reviewedAt": "2026-09-14T00:43:53.625753Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "GeoVAD-Bench diagnoses visual chain-of-thought for geometry problem solving along five dimensions: perception, auxiliary quality, utilization, deductive reasoning, and final correctness, with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings.", "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To bridge this gap, we introduce GeoVAD-Bench, a diagnostic benchmark that pairs a fine-grained five-dimensional trajectory diagnosis covering perception, auxiliary quality, utilization, deductive reasoning, and final correctness with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings to systematically isolate intermediate error modes, the causal gains of visual aids, and the resulting autonomy gap."}, "familyId": "bmf_4b79b1098a6a", "firstSeenAt": "2026-09-14", "heat": null, "id": "bm_geovad-bench_dadd2d8d", "indexedAt": "2026-09-14T00:39:56.575498Z", "industrySectors": ["Robotics"], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.12606", "pdf": "https://arxiv.org/pdf/2609.12606", "project": null, "report": "https://arxiv.org/abs/2609.12606"}, "metrics": [], "motivation": "While multimodal reasoning has advanced rapidly, solving complex geometry problems critically hinges on active visual assistance, such as constructing auxiliary lines, spurring the rise of Visual Chain-of-Thought (VCoT).", "name": "GeoVAD-Bench", "oneLine": "GeoVAD-Bench diagnoses visual chain-of-thought for geometry problem solving along five dimensions: perception, auxiliary quality, utilization, deductive reasoning, and final correctness, with controlled No-Aux, Auto-Aux, and GT-Aux intervention settings.", "paperTitle": "Beyond Generation and Accuracy: Diagnosing and Enhancing Visual Chain-of-Thought for Geometry Problem Solving", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.12606", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-14T00:39:56.575498Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.3, "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": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.3871870587422144, "value": 7}}, "confidence": "Low", "coverage": 0.3, "rank": null, "score": 47}, "method": "attention-ranking-v13", "rank": null, "score": 47}, "90d": {"confidence": "Low", "coverage": 0.15, "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": 0.3674261111536563, "value": 7}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 48}, "method": "attention-ranking-v13", "rank": null, "score": 48}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-11", "source": {"authors": ["Zhitong Dong", "Jicai Pan", "Yingguo Gao", "Jingting Ding", "Hao Chen", "Jinjie Gu"], "categories": ["cs.AI"], "id": "2609.12606", "title": "Beyond Generation and Accuracy: Diagnosing and Enhancing Visual Chain-of-Thought for Geometry Problem Solving", "type": "arxiv", "url": "https://arxiv.org/abs/2609.12606"}, "sourceUpdatedAt": "2026-09-11T09:00:57Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Geometry solving evaluations often report only final accuracy; a trajectory-level diagnosis with controlled interventions can separate intermediate error modes from outcome errors and clarify the value of visual aids."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
