[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-researchers-let-trading-bots-rewrite-their-own-instructions":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},6477,"researchers-let-trading-bots-rewrite-their-own-instructions","Researchers Let Trading Bots Rewrite Their Own Instructions","A new framework called EvolveTrade lets an AI trading agent edit its own system prompt after every trading round, using past results as feedback.","A new AI framework lets trading bots rewrite their own strategy instructions after every round of trades.\n\nResearchers built EvolveTrade, which treats an LLM trading agent's system prompt as an adjustable policy rather than a fixed script. After each trading interval, a separate Policy Agent reviews the trading agent's decision history and portfolio results, then rewrites the instructions guiding how it gathers evidence, calls tools, checks signals, and manages risk. The underlying LLM itself never changes - only the text-based playbook it follows. Tested across multiple market regimes and two different LLM backbones, the self-updating agents beat fixed-policy versions on Sharpe ratio and cumulative return in most setups.\n\nMost LLM trading agents ship with a strategy baked in at launch and never adjust it, even as market conditions shift. EvolveTrade's approach - improving the instructions rather than retraining the model - is cheap compared to fine-tuning, and the researchers found the evolved policies leaned more on code-based analysis and activated computations suited to the current market regime. That is a meaningful shift: the system is learning how to think about a trade, not just what to predict.\n\nWhether that adaptability survives contact with real markets, where slippage and liquidity punish overconfidence, is the harder question backtests alone cannot answer.","[\"ai\",\"algorithmic-trading\",\"llm-agents\",\"research\"]","2026-09-17T04:00:00.000Z","2026-09-17T20:04:22.647Z","2026-09-17T20:04:34.567Z","published",null,[],"ai",[24,26,27,28],"algorithmic-trading","llm-agents","research",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.17632",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",3853,"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"]