{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.16753", "kind": "paper", "url": "https://arxiv.org/abs/2606.16753"}], "authors": ["Rafael Ferreira", "Inês Vieira", "Inês Calvo", "James Furtado", "Iago Paulo", "Diogo Tavares", "Diogo Glória-Silva", "David Semedo", "João Magalhães"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.16753", "source_url": "https://arxiv.org/abs/2606.16753"}, "categories": ["General AI", "Safety & Trustworthiness", "cs.CL"], "description": {"en": "P3B3 is an expert-curated benchmark of conversational prompts for evaluating variety bias and controllability in LLMs across European and Brazilian Portuguese."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.16753", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.16753", "title": "P3B3"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Laboratório para a Ciência da Computação e Informática", "Universidade Nova de Lisboa"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.16753", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.16753", "source_url": "https://openalex.org/W7164917128"}, "key": "claire-radar:2606.16753", "modality": null, "name": "P3B3", "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": "unknown", "origin_record_id": "bm_p3b3_5a9c84e4", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "cdc39997132a094785452e190b273273cb7c5aed7a42de485090d8bc2fe5de8d", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.16753"}, "publisher": null, "released": "2026-06-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.16753", "source_url": "https://arxiv.org/abs/2606.16753"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-16753", "source": "claire_radar", "source_benchmark_id": "2606.16753", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Safety & Trustworthiness", "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": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.16753", "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", "lastSuccessfulDate": "2026-10-01", "state": "stale"}}, "status": "stale"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "59b0a456a8458ea20c319288e400dc669f34c77ec29b3087ec09ef53886424ea", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "P3B3 is an expert-curated benchmark of conversational prompts for evaluating variety bias and controllability in LLMs across European and Brazilian Portuguese.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence"], "snippet": "To address this gap, we introduce P3B3, an expert-curated language variety agnostic benchmark of conversational prompts, along with an evaluation framework for measuring variety bias and controllability."}, "familyId": "bmf_35600c5fa04e", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_p3b3_5a9c84e4", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.16753", "paper": "https://arxiv.org/abs/2606.16753", "pdf": "https://arxiv.org/pdf/2606.16753", "project": null, "report": "https://arxiv.org/abs/2606.16753"}, "metrics": [], "motivation": "As Large Language Models (LLMs) become embedded in everyday communication, capturing regional linguistic variation is essential for reliable and equitable language use.", "name": "P3B3", "oneLine": "P3B3 is an expert-curated benchmark of conversational prompts for evaluating variety bias and controllability in LLMs across European and Brazilian Portuguese.", "paperTitle": "P3B3: A Multi-Turn Conversational Benchmark for Measuring European and Brazilian Portuguese Variety Bias in LLMs", "primaryDomain": "General AI", "publication": {"evidence": "Accepted at MeLLM Workshop at ACL 2026", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.16753", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "MeLLM Workshop at ACL 2026", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-15", "source": {"authors": ["Rafael Ferreira", "In\\^es Vieira", "In\\^es Calvo", "James Furtado", "Iago Paulo", "Diogo Tavares", "Diogo Gl\\'oria-Silva", "David Semedo", "Jo\\~ao Magalh\\~aes"], "categories": ["cs.CL", "cs.AI", "cs.LG"], "id": "2606.16753", "title": "P3B3: A Multi-Turn Conversational Benchmark for Measuring European and Brazilian Portuguese Variety Bias in LLMs", "type": "arxiv", "url": "https://arxiv.org/abs/2606.16753"}, "sourceUpdatedAt": "2026-06-15T14:10:19Z", "topics": ["cs.CL"], "venueAttempts": [{"decisionRaw": "Accepted at MeLLM Workshop at ACL 2026", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at MeLLM Workshop at ACL 2026", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.16753"}], "reviewStatus": "accepted", "venueName": "MeLLM Workshop at ACL 2026"}], "whyItMatters": "Evaluates regional variety bias in Portuguese LLMs, addressing underrepresentation and controllability gaps."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
