[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-poisoned-rag-contexts-drop-llama-31-accuracy-from-779-to-435":10,"sections":45},{"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":34,"tags":35,"sources":40,"feedback":44,"feedback_at":22,"cost_usd":44,"total_tokens":44},6384,"poisoned-rag-contexts-drop-llama-31-accuracy-from-779-to-435","Poisoned RAG Contexts Drop Llama 3.1 Accuracy From 77.9% to 43.5%","A new study finds a small quantized Llama 3.1 model's fact-checking accuracy plunges from 77.9% to 43.5% when all three retrieved passages are poisoned.","Feed a retrieval-augmented model bad documents, and it believes them almost half the time.\n\nResearchers ran Llama 3.1 8B, a small quantized model, through a fact-checking task built from the FEVER dataset. They tested three ways to poison retrieved passages: swapping entities, swapping numbers, and inverting claims with negation. Across a sweep of 588 runs, they corrupted zero, one, two, or three of the three passages the model retrieved before answering. Accuracy dropped from 77.9% on clean context to 43.5% when every retrieved passage was tampered with.\n\nThe model didn't mostly invent new lies. Its main response to poisoned context was abstaining rather than fabricating, and a rough proxy for unsupported claims actually fell under attack rather than rose. That's a partial defense, but abstaining on a fact-check is still a wrong answer, and entity swaps flipped more correct answers than any other tactic tested.\n\nThe paper is small-scale and its automated grading is coarse, so treat the entity-versus-number contrast as a lead rather than a verdict - but the headline number stands on its own: corrupt every document a model retrieves, and it gets fact-checks right less than half the time.","[\"rag\",\"ai-security\",\"llama\",\"research\"]","2026-09-11T04:00:00.000Z","2026-09-11T09:50:18.641Z","2026-09-11T09:50:30.542Z","published",null,[24,30],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"Attribute the findings to the actual source (arXiv:2609.09243) with a link or paper title instead of the unnamed 'researchers' — right now there's no institution, publication, or link anywhere in the body.","resolved",{"id":31,"reviewer":26,"round":32,"reason":33,"status":29},"editor-r2",2,"Make the headline's numbers (78%\u002F44%) match the precise 77.9%\u002F43.5% figures used in the dek and body instead of rounding them differently.","ai",[36,37,38,39],"rag","ai-security","llama","research",[41],{"name":42,"url":43},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.09243",0,{"sections":46},[47,50,54,58,63,68,73,76,81,85,90,95,100,105],{"name":48,"slug":34,"count":49,"latest_published_at":18},"AI",3543,{"name":51,"slug":52,"count":53,"latest_published_at":18},"Security","security",637,{"name":55,"slug":56,"count":57,"latest_published_at":18},"Policy","policy",338,{"name":59,"slug":60,"count":61,"latest_published_at":62},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":64,"slug":65,"count":66,"latest_published_at":67},"Hardware","hardware",153,"2026-09-09T15:12:32.000Z",{"name":69,"slug":70,"count":71,"latest_published_at":72},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":74,"slug":75,"count":71,"latest_published_at":18},"Science","science",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":18},"Dev Tools","dev-tools",70,{"name":86,"slug":87,"count":88,"latest_published_at":89},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":106,"slug":107,"count":108,"latest_published_at":109},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]