{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.01059", "kind": "paper", "url": "https://arxiv.org/abs/2609.01059"}], "authors": ["Jiayu Ding", "Zhuodong Liu", "Lei Zhang", "Manyu Xiong", "Hongbo Jin", "Haoran Tang", "Hongbo Zhang", "Changen Zhu", "Wenbo Xing"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.01059", "source_url": "https://arxiv.org/abs/2609.01059"}, "categories": ["General AI", "Geometric reasoning", "Multimodal"], "description": {"en": "Evaluates vision-language models on ego-motion understanding using counterfactual 3D rendering to separate visual appearance from kinematic properties."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.01059", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.01059", "title": "Dyn-3D"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.01059", "modality": null, "name": "Dyn-3D", "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": "true", "origin_record_id": "bm_dyn-3d_1d5a4006", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "2d39c23990de8939d180b50cda881f516744d9e9692ba633e1b9ca5da30ed62d", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-03T00:59:34.317835Z", "source_url": "https://arxiv.org/abs/2609.01059"}, "publisher": null, "released": "2026-09-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.01059", "source_url": "https://arxiv.org/abs/2609.01059"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-01059", "source": "claire_radar", "source_benchmark_id": "2609.01059", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "attention": {"asOf": "2026-09-02", "attemptedAt": "2026-09-02T06:30:32.214812Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.01059", "observedAt": "2026-08-31T04:56:28.279469Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-02T06:30:32.214812Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "The benchmark tackles a niche ambiguity in vision-language models with no supplied code or project page, suggesting limited early attention.", "score": 40}, "capabilities": ["Geometric reasoning"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-03T00:59:34.317835Z", "inputHash": "1eccadcf71911a33f9f8fe61cbaaf48c954b004d6faf63cc4cecd9186f8dd408", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "we introduce Dyn-3D, a benchmark using counterfactual 3D rendering", "canonicalNameSource": "abstract", "decisionReason": "Abstract formally introduces the benchmark with a clear evaluation protocol and implies release for reuse.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-03T00:59:34.317835Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates vision-language models on ego-motion understanding using counterfactual 3D rendering to separate visual appearance from kinematic properties.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To evaluate this, we introduce Dyn-3D, a benchmark using counterfactual 3D rendering to rigorously decouple visual changes from true kinematic properties."}, "familyId": "bmf_d1fcbce846bd", "firstSeenAt": "2026-09-02", "heat": null, "id": "bm_dyn-3d_1d5a4006", "indexedAt": "2026-09-02T05:21:06.631990Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.01059", "pdf": "https://arxiv.org/pdf/2609.01059", "project": null, "report": "https://arxiv.org/abs/2609.01059"}, "metrics": [], "motivation": "As Vision-Language Models (VLMs) tackle dynamic 3D spatial reasoning, ego-motion perception becomes essential to resolve monocular scale ambiguity.", "name": "Dyn-3D", "oneLine": "Evaluates vision-language models on ego-motion understanding using counterfactual 3D rendering to separate visual appearance from kinematic properties.", "paperTitle": "Dyn-3D: Unveiling and Resolving Ego-Motion Ambiguity in Vision-Language Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.01059", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-02T05:21:06.631990Z"}, "publications": [], "ranking": {"30d": {"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": 30}, "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}, "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}, "today": {"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": 2, "historyDate": "2026-08-31", "rank": null, "score": null, "windowDays": 1}, "level": {"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}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.3, "role": "bonus", "value": 30}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": 8}, "method": "attention-ranking-v13", "rank": null, "score": 8}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-01", "source": {"authors": ["Jiayu Ding", "Zhuodong Liu", "Lei Zhang", "Manyu Xiong", "Hongbo Jin", "Haoran Tang", "Hongbo Zhang", "Changen Zhu", "Wenbo Xing"], "categories": ["cs.CV"], "id": "2609.01059", "title": "Dyn-3D: Unveiling and Resolving Ego-Motion Ambiguity in Vision-Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2609.01059"}, "sourceUpdatedAt": "2026-09-01T10:54:27Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Addresses a key gap in vision-language evaluation by testing whether models infer true motion rather than relying on spurious visual cues."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
