{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.03803", "kind": "paper", "url": "https://arxiv.org/abs/2608.03803"}, {"kind": "website", "url": "https://m-gate.ai"}], "authors": ["Tomáš Burkert", "Angelika Peljak-Łapińska", "David Zelený"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.03803", "source_url": "https://arxiv.org/abs/2608.03803"}, "categories": ["Language & Knowledge", "Software & AI Compute", "cs.CL"], "description": {"en": "A multilingual proficiency benchmark covering grammatical error detection, round-trip translation, and tokenizer efficiency across 30 languages."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.03803", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.03803", "title": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.03803", "modality": null, "name": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models", "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": "2026-09-25T04:40:22Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_m-gate-multilingual-grammar-accuracy-in-tr_41f20366", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "43cb800951f0b6b70f9937b6d429f99bbd8ddb7c1fac17dead34268d3594a779", "relation": "introduces", "review_model": "codex-cli-default", "review_state": "ai-reviewed", "reviewed_at": "2026-09-24T17:36:59.562667Z", "source_url": "https://arxiv.org/abs/2608.03803"}, "publisher": null, "released": "2026-08-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.03803", "source_url": "https://arxiv.org/abs/2608.03803"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-03803", "source": "claire_radar", "source_benchmark_id": "2608.03803", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Language & Knowledge", "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "Broad multilingual coverage, evaluation of more than 50 models, and a public leaderboard are likely to generate substantial near-term interest.", "score": 82}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-24T17:36:59.562667Z", "inputHash": "c2a24ac8d2a61691b88c0653c6c831599b4efc07bd7624610443fa6f28bd2258", "model": "codex-cli-default"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models", "canonicalNameSource": "paper_title", "decisionReason": "M-GATE is formally named, defines concrete tasks and scoring, and offers a public leaderboard as a continuing submission path.", "evaluationEvidence": {"reuse": "test items are kept private behind a continuously updated public leaderboard", "scoring": "scored by a three-provider LLM judge panel validated against professional annotators", "task": "grammatical error detection"}, "model": "codex-cli-default", "reviewedAt": "2026-09-24T17:36:59.562667Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A multilingual proficiency benchmark covering grammatical error detection, round-trip translation, and tokenizer efficiency across 30 languages.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce M-GATE (Multilingual Grammar, Accuracy in Translation, and Efficiency), a benchmark of linguistic proficiency spanning 30 typologically diverse languages from high- to low-resource."}, "familyId": "bmf_4e9d23cd0a24", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_m-gate-multilingual-grammar-accuracy-in-tr_41f20366", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2608.03803", "pdf": "https://arxiv.org/pdf/2608.03803", "project": "https://m-gate.ai", "report": "https://arxiv.org/abs/2608.03803"}, "metrics": [], "motivation": "Multilingual language models are deployed across a hundred or more languages, yet most benchmarks test whether a model can perform a task _in_ a language rather than whether it commands the language itself, conflating fluency with proficiency.", "name": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models", "oneLine": "A multilingual proficiency benchmark covering grammatical error detection, round-trip translation, and tokenizer efficiency across 30 languages.", "paperTitle": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.03803", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "readiness": "Paper only", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-08-04", "reviewLifecycle": {"state": "pending"}, "source": {"authors": ["Tom\\'a\\v{s} Burkert", "Angelika Peljak-{\\L}api\\'nska", "David Zelen\\'y"], "categories": ["cs.CL", "cs.LG"], "id": "2608.03803", "title": "M-GATE: Multilingual Grammar, Accuracy in Translation, and Efficiency Benchmark for Large Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2608.03803"}, "sourceUpdatedAt": "2026-08-04T15:16:05Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "It separates linguistic proficiency from surface fluency and measures performance across typologically diverse, including low-resource, languages."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
