{"schema_version":"papers/paper-detail-v1","title":"Can Vision Language Models Judge Action Quality? An Empirical Evaluation","surface":"papers","opportunity_kernel":{"paper_id":"68b3b19e-34e1-4ed2-86ce-26d0c6037b4e","title":"Can Vision Language Models Judge Action Quality? An Empirical Evaluation","authors":["Miguel Monte e Freitas","Rui Henriques","Ricardo Rei","Pedro Henrique Martins"],"arxiv_id":"2604.08294v1","doi":null,"published_at":"2026-04-09T14:29:19.000Z","score_object":{"overall":{"value":2,"scale":"0-10","confidence":0.85,"confidence_reason":"Backfilled from persisted papers.viability_score.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:23:10.479Z","fresh_until":"2026-05-10T17:23:10.479Z","is_stale":true,"source_count":1,"missingness":[]},"technical":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No persisted technical score source was available; marked low confidence.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:23:10.479Z","fresh_until":"2026-04-24T17:23:10.479Z","is_stale":true,"source_count":0,"missingness":["reproducibility_results.reproducibility_score","deployability_scores.score","paper_extraction_scorecards.reconstruction_score"]},"commercial":{"value":0,"scale":"0-10","confidence":0.75,"confidence_reason":"Backfilled from persisted commercial_flags and repo availability.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:23:10.479Z","fresh_until":"2026-05-10T17:23:10.479Z","is_stale":true,"source_count":1,"missingness":[]},"market":{"value":1,"scale":"0-10","confidence":0.45,"confidence_reason":"Heuristic market score materialized from paper metadata, repo availability, and deployability.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:42:52.164Z","fresh_until":"2026-04-24T17:42:52.164Z","is_stale":true,"source_count":1,"missingness":[]},"team":{"value":3.6,"scale":"0-10","confidence":0.28,"confidence_reason":"Heuristic fallback from paper author count.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:23:10.479Z","fresh_until":"2026-04-24T17:23:10.479Z","is_stale":true,"source_count":1,"missingness":["paper_author_affiliations"]},"methodology":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No persisted methodology score source was available; marked low confidence.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-04-10T17:23:10.479Z","fresh_until":"2026-05-10T17:23:10.479Z","is_stale":true,"source_count":0,"missingness":["paper_extraction_scorecards.total_score","paper_extraction_scorecards.standard_extraction_score"]}},"evidence_receipt":{"freshness":"stale","proof_status":"unverified","repo_status":"missing","references_count":0,"source_count":3,"coverage":0.3333,"missingness":["repo_url","references","proof_status","paper_extraction_scorecards"],"unresolved_unknowns":["proof verification has not been recorded yet"],"last_verification_at":"2026-04-10T17:42:52.164Z"},"lineage_hash":"8f60eecaa4941f6b60ab98041cd63a84f69eb535d49d59073558a467f7b8153f"},"distribution":null,"replication_evidence":[],"author_dna":[]}