{"id":4670,"date":"2025-06-18T04:03:44","date_gmt":"2025-06-18T04:03:44","guid":{"rendered":"https:\/\/mailitics.com\/index.php\/2025\/06\/18\/lost-in-the-middle-how-llm-architecture-and-training-data-shape-ais-position-bias\/"},"modified":"2025-06-18T04:03:44","modified_gmt":"2025-06-18T04:03:44","slug":"lost-in-the-middle-how-llm-architecture-and-training-data-shape-ais-position-bias","status":"publish","type":"post","link":"https:\/\/mailitics.com\/index.php\/2025\/06\/18\/lost-in-the-middle-how-llm-architecture-and-training-data-shape-ais-position-bias\/","title":{"rendered":"Lost in the middle: How LLM architecture and training data shape AI&#8217;s position bias"},"content":{"rendered":"\n<div>Lost in the middle: How LLM architecture and training data shape AI&#8217;s position bias<\/div>\n<p> \t<BR><br \/>\n<BR><\/BR><br \/>\n    <!-- no image --><br \/>\n \t<BR><br \/>\n<BR><\/BR><\/p>\n<div>Research has shown that large language models (LLMs) tend to overemphasize information at the beginning and end of a document or conversation, while neglecting the middle.<\/div>\n<p> \t<BR><br \/>\n <BR><\/BR><\/p>\n<p> \t<BR><br \/>\n<BR><\/BR><br \/>\n<a href=\"https:\/\/techxplore.com\/news\/2025-06-lost-middle-llm-architecture-ai.html\">Go to techxplore<\/a><br \/>\n \t<BR><br \/>\n <BR><\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Lost in the middle: How LLM architecture and training data shape AI&#8217;s position bias Research has shown that large language models (LLMs) tend to overemphasize information at the beginning and end of a document or conversation, while neglecting the middle. Go to techxplore<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[54,45],"tags":[50],"class_list":["post-4670","post","type-post","status-publish","format-standard","hentry","category-computer-sciences","category-techxplore","tag-techxplore"],"_links":{"self":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/4670"}],"collection":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/comments?post=4670"}],"version-history":[{"count":0,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/4670\/revisions"}],"wp:attachment":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/media?parent=4670"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/categories?post=4670"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/tags?post=4670"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}