{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.28218", "kind": "paper", "url": "https://arxiv.org/abs/2605.28218"}, {"kind": "repo", "url": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench"}], "authors": ["Mingrui Sun", "Mao Zheng", "Zheng Li", "Mingyang Song"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.28218", "source_url": "https://arxiv.org/abs/2605.28218"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "Evaluates multilingual translation systems on instruction following across seven languages, six constraint types, and compositional multi-constraint requests."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.28218", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.28218", "title": "IFMTBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.28218", "modality": null, "name": "IFMTBench", "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_ifmtbench_b83f8c4f", "origin_source": "arxiv", "recognition_confidence": "0.95", "record_sha256": "1032e33e0e84936b38c6e19654629daca5a4802f6264c438130eeefd48d605e3", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.28218"}, "publisher": null, "released": "2026-05-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.28218", "source_url": "https://arxiv.org/abs/2605.28218"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-28218", "source": "claire_radar", "source_benchmark_id": "2605.28218", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Tencent-Hunyuan/Hy-MT2", "githubScope": "hosting_repo", "githubStars": 676, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2605.28218", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "fresh"}, "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": "8ffde504a4f6f0125b70268f3de10201211ac5b384be2b71235cdc817cda7118", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates multilingual translation systems on instruction following across seven languages, six constraint types, and compositional multi-constraint requests.", "domainCuration": {"method": "primary-source-review-v1", "state": "reviewed"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "We introduce \\bench, a benchmark for multilingual translation instruction following covering seven languages, with 4,506 single-constraint and 2,838 multi-constraint items spanning six constraint dimensions and five compositional patterns with instructions issued in all seven languages."}, "familyId": "bmf_c2644d034b9f", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench", "url": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/Tencent-Hunyuan/Hy-MT2", "url": "https://huggingface.co/papers/2605.28218"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench", "url": "https://arxiv.org/abs/2605.28218"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench.", "url": "https://arxiv.org/abs/2605.28218"}], "scope": "hosting_repo", "url": "https://github.com/Tencent-Hunyuan/Hy-MT2", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_ifmtbench_b83f8c4f", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench", "data": null, "hfPaper": "https://huggingface.co/papers/2605.28218", "paper": "https://arxiv.org/abs/2605.28218", "pdf": "https://arxiv.org/pdf/2605.28218", "project": null, "report": "https://arxiv.org/abs/2605.28218"}, "metrics": [], "motivation": "Modern translation workflows demand more than semantic equivalence.", "name": "IFMTBench", "oneLine": "Evaluates multilingual translation systems on instruction following across seven languages, six constraint types, and compositional multi-constraint requests.", "paperTitle": "IFMTBench: A Comprehensive Benchmark for Multilingual Translation Instruction Following", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.28218", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Tencent Hunyuan", "organizationType": "company-research-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Tencent-Hunyuan/Hy-MT2/tree/main/IFMTBench"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 0.95, "relation": "introduces", "releasedAt": "2026-05-27", "source": {"authors": ["Mingrui Sun", "Mao Zheng", "Zheng Li", "Mingyang Song"], "categories": ["cs.CL"], "id": "2605.28218", "title": "IFMTBench: A Comprehensive Benchmark for Multilingual Translation Instruction Following", "type": "arxiv", "url": "https://arxiv.org/abs/2605.28218"}, "sourceUpdatedAt": "2026-05-27T09:35:27Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Shows whether translation systems can follow terminology, style, length, audience, and other workflow constraints while preserving translation quality."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
