{"record": {"aliases": ["Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEval"], "artifacts": [{"id": "arxiv:2609.11172", "kind": "paper", "url": "https://arxiv.org/abs/2609.11172"}, {"id": "gh:internrobotics/egogeneval", "kind": "repo", "url": "https://github.com/InternRobotics/EgoGenEval"}], "authors": ["Yilin Long", "Chenming Zhu", "Zitang Gou", "Jingli Lin", "Tai Wang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.11172", "source_url": "https://arxiv.org/abs/2609.11172"}, "categories": ["Geometric reasoning", "Robotics & Embodied AI", "Vision & 3D", "cs.CV"], "description": {"en": "EgoGenEval offers 1,400 cases and 2,360 target views to assess physical consistency of visual generators under ego-motion, measuring Camera Motion Grounding and Scene State Preservation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.11172", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.11172", "title": "EgoGenEval"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.11172", "modality": null, "name": "EgoGenEval", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "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_beyond-visual-quality-evaluating-physical-_fca65bd7", "origin_source": "arxiv", "recognition_confidence": "0.55", "record_sha256": "042e3eb429078682f998cb525ff180e51d38da893593856e3447d1d1f82b88a6", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-11T06:25:08.028122Z", "source_url": "https://arxiv.org/abs/2609.11172"}, "publisher": null, "released": "2026-09-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.11172", "source_url": "https://arxiv.org/abs/2609.11172"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-11172", "source": "claire_radar", "source_benchmark_id": "2609.11172", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEval"], "annotation": "Unknown", "applicationDomains": ["Robotics & Embodied AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-09-11", "attemptedAt": "2026-09-11T01:40:00.759237Z", "githubRepo": "https://github.com/InternRobotics/EgoGenEval", "githubScope": "benchmark_repo", "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.11172", "observedAt": "2026-09-10T12:24:43.080049Z", "signalStatus": {"githubStars": {"checkedAt": "2026-09-11T01:40:00.759237Z", "state": "unavailable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-11T01:40:00.759237Z", "state": "unavailable"}}, "status": "unavailable"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark targets an emergent problem in generative visual models and includes released code, likely drawing attention from computer vision and generation communities.", "score": 60}, "capabilities": ["Geometric reasoning"], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-11T06:25:08.028122Z", "inputHash": "1f94e0129a44581460109e91557abb780fc09d6f7fbfb26c997d5571321eb067", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "We introduce EgoGenEval", "canonicalNameSource": "abstract", "decisionReason": "The paper explicitly introduces a named benchmark with defined metrics, validation, and a public GitHub code repository, indicating a reusable evaluation protocol.", "evaluationEvidence": {"reuse": "code: https://github.com/InternRobotics/EgoGenEval", "scoring": "It separately measures Camera Motion Grounding (CMG) and Scene State Preservation (SSP), with both metrics validated against blinded human judgments", "task": "1,400 cases and 2,360 target views spanning single-step and multi-step ego-motion"}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-11T06:25:08.028122Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "EgoGenEval offers 1,400 cases and 2,360 target views to assess physical consistency of visual generators under ego-motion, measuring Camera Motion Grounding and Scene State Preservation.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "We introduce EgoGenEval, a geometry-grounded, pose-free benchmark designed to evaluate the physical consistency of visual generators under ego-motion, and organize our study into two parts."}, "familyId": "bmf_2e06dd6c9f72", "firstSeenAt": "2026-09-11", "heat": null, "id": "bm_beyond-visual-quality-evaluating-physical-_fca65bd7", "indexedAt": "2026-09-11T00:48:15.671466Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/InternRobotics/EgoGenEval", "data": null, "paper": "https://arxiv.org/abs/2609.11172", "pdf": "https://arxiv.org/pdf/2609.11172", "project": null, "report": "https://arxiv.org/abs/2609.11172"}, "metrics": [], "motivation": "Recent visual generators produce high-fidelity images yet often violate physical consistency under ego-motion, limiting their use for spatial reasoning and embodied planning.", "name": "EgoGenEval", "oneLine": "EgoGenEval offers 1,400 cases and 2,360 target views to assess physical consistency of visual generators under ego-motion, measuring Camera Motion Grounding and Scene State Preservation.", "paperTitle": "Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEval", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.11172", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-11T00:48:15.671466Z"}, "publications": [], "publishers": [{"name": "InternRobotics", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/InternRobotics/EgoGenEval"}], "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": 1, "historyDate": "2026-09-10", "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": 0.72, "role": "bonus", "value": 72}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": 18}, "method": "attention-ranking-v13", "rank": null, "score": 18}}, "readiness": "Runnable", "recognitionConfidence": 0.55, "relation": "introduces", "releasedAt": "2026-09-10", "source": {"authors": ["Yilin Long", "Chenming Zhu", "Zitang Gou", "Jingli Lin", "and Tai Wang"], "categories": ["cs.CV", "cs.AI"], "id": "2609.11172", "title": "Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEval", "type": "arxiv", "url": "https://arxiv.org/abs/2609.11172"}, "sourceUpdatedAt": "2026-09-10T07:19:50Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Physical consistency under ego-motion is underexplored in visual generation, and this benchmark enables comparing models on both camera grounding and temporal scene preservation for embodied planning."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
