{"record": {"aliases": ["One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning"], "artifacts": [{"id": "arxiv:2609.05260", "kind": "paper", "url": "https://arxiv.org/abs/2609.05260"}], "authors": ["Yiwei Liu", "Luwei Yang", "Shunbo Lei"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.05260", "source_url": "https://arxiv.org/abs/2609.05260"}, "categories": ["Reasoning", "Robotics", "Robotics & Embodied AI"], "description": {"en": "The paper describes a real-robot benchmark evaluating language-conditioned embodied reasoning with instruction pairs, but the benchmark name is not formally declared in the abstract; it uses the phrase 'introduce One Word, Different Action' without explicit benchmark declaration."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.05260", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.05260", "title": "One Word, Different Action"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Chinese University of Hong Kong", "Chinese University of Hong Kong, Shenzhen", "Shenzhen Research Institute of Big Data"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2609.05260", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2609.05260", "source_url": "https://openalex.org/W7211896370"}, "key": "claire-radar:2609.05260", "modality": null, "name": "One Word, Different Action", "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_one-word-different-action-a-real-robot-ben_140e4fe9", "origin_source": "arxiv", "recognition_confidence": "0.6", "record_sha256": "de3dceb4b0cc2b6c462b2ff073bcf3d2da726ac696b78cf0a3c85340701869c0", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-08T00:56:29.478814Z", "source_url": "https://arxiv.org/abs/2609.05260"}, "publisher": null, "released": "2026-09-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.05260", "source_url": "https://arxiv.org/abs/2609.05260"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-05260", "source": "claire_radar", "source_benchmark_id": "2609.05260", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning"], "annotation": "Unknown", "applicationDomains": ["Robotics & Embodied AI"], "area": "Robotics & Embodied AI", "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/2609.05260", "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": "Insufficient evidence for a formal benchmark release, so attention forecast is minimal.", "score": 0}, "capabilities": ["Reasoning"], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-08T08:59:06.803818Z", "inputHash": "d259e41ab6128320d5586ea704e1266d6f45131581db7a5cb0652fee0ab0973e", "model": "deepseek-v4-flash"}, "curation": {"canonicalNameEvidence": "One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning", "canonicalNameSource": "paper_title", "decisionReason": "No public dataset, code, or evaluation platform is referenced; the benchmark lacks a demonstrated path for other teams to run it.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-08T00:56:29.478814Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "The paper describes a real-robot benchmark evaluating language-conditioned embodied reasoning with instruction pairs, but the benchmark name is not formally declared in the abstract; it uses the phrase 'introduce One Word, Different Action' without explicit benchmark declaration.", "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce One Word, Different Action, a real-robot benchmark built on physical decision states and executable actions, using task-preserving and task-changing instruction pairs to jointly evaluate Decision Invariance and Decision Sensitivity, with further evaluation under multi-constraint reasoning and real-RGB grounding."}, "familyId": "bmf_22dfc6d55a04", "firstSeenAt": "2026-09-07", "heat": null, "id": "bm_one-word-different-action-a-real-robot-ben_140e4fe9", "indexedAt": "2026-09-07T06:22:02.605340Z", "industrySectors": ["Robotics"], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.05260", "pdf": "https://arxiv.org/pdf/2609.05260", "project": null, "report": "https://arxiv.org/abs/2609.05260"}, "metrics": [], "motivation": "Natural-language instruction changes can directly alter robot behavior.", "name": "One Word, Different Action", "oneLine": "The paper describes a real-robot benchmark evaluating language-conditioned embodied reasoning with instruction pairs, but the benchmark name is not formally declared in the abstract; it uses the phrase 'introduce One Word, Different Action' without explicit benchmark declaration.", "paperTitle": "One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.05260", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-07T06:22:02.605340Z"}, "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": 0.6, "relation": "introduces", "releasedAt": "2026-09-04", "source": {"authors": ["Yiwei Liu", "Luwei Yang", "Shunbo Lei"], "categories": ["cs.RO"], "id": "2609.05260", "title": "One Word, Different Action: A Real-Robot Benchmark for Language-Conditioned Embodied Reasoning", "type": "arxiv", "url": "https://arxiv.org/abs/2609.05260"}, "sourceUpdatedAt": "2026-09-04T15:21:58Z", "topics": ["Robotics", "Reasoning"], "venueAttempts": [], "whyItMatters": "Potentially addresses the gap in evaluating how embodied models handle instruction changes, but unclear benchmark naming and lack of explicit details hamper standalone reuse."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
