[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-benchmark-tests-cultural-value-leanings-in-ai-models":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},6616,"new-benchmark-tests-cultural-value-leanings-in-ai-models","New Benchmark Tests Cultural Value Leanings in AI Models","A new benchmark called GYROval tests whether large language models hold stable, translation-proof cultural value leanings across two classic axes.","A new benchmark scores how large language models lean on two classic dimensions of cultural values, and the rankings hold up even when you translate the questions or crank up the randomness.\n\nResearchers built GYROval to measure where LLMs fall on the Inglehart-Welzel cultural value axes across multiple domains and roles. Like the earlier CDEval benchmark, its test items are binary contrastive scenarios: two legitimate options, no correct answer, and a model's score is simply how often it picks one pole over the other. Twenty models were tested this way. Eleven of those twenty were also run against a Russian translation of the identical items and at a second sampling temperature, then checked for consistency using a tie-corrected Kendall's W statistic against an empirical permutation null.\n\nThis matters because LLMs now field questions with real cultural stakes, on family, authority, gender roles, and religion, and their answers reflect training data and fine-tuning choices, not objective facts. A benchmark that survives translation and temperature changes gives researchers a way to argue a model's cultural lean is a real, measurable property rather than noise from how a question happens to be phrased.\n\nIt is a measurement tool, not a verdict: the paper does not say which labs' models skew traditional or secular-rational, leaving the more pointed question, and probably several follow-up papers, for later.","[\"ai\",\"benchmarks\",\"llm evaluation\",\"culture\"]","2026-09-17T04:00:00.000Z","2026-09-18T02:46:24.518Z","2026-09-18T02:46:36.416Z","published",null,[],"ai",[24,26,27,28],"benchmarks","llm evaluation","culture",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.18384",0,{"sections":35},[36,40,44,49,54,58,62,67,72,76,81,86,91,96],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",3852,"2026-09-17T08:27:09.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",648,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",338,"2026-09-11T04:00:00.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":18},"Hardware","hardware",154,{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",114,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":71},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":73,"slug":74,"count":75,"latest_published_at":18},"Dev Tools","dev-tools",73,{"name":77,"slug":78,"count":79,"latest_published_at":80},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]