Public articles linked to the same research event.
arXiv Studying end-to-end search with 12 agent models across three domains, the work finds each model prefers some sources and avoids others with broad agreement on which; an item satisfying one requirement fewer is selected about two-thirds of the time when it comes from a preferred source while the better item comes from a dispreferred source, but almost never in the reverse case; hiding source information weakens the preference and relabeling an item with a preferred source raises its selection rate; training that rewards better items can make a source a shortcut for requirement satisfaction, and missing information can trigger preconceptions about the source, while supplying missing information or a prompt countering those preconceptions reduces source preference.
Studying end-to-end search with 12 agent models across three domains, the work finds each model prefers some sources and avoids others with broad agreement on which; an item satisfying one requirement fewer is selected about two-thirds of the time when it comes from a preferred source while the better item comes from a dispreferred source, but almost never in the reverse case; hiding source information weakens the preference and relabeling an item with a preferred source raises its selection rate; training that rewards better items can make a source a shortcut for requirement satisfaction, and missing information can trigger preconceptions about the source, while supplying missing information or a prompt countering those preconceptions reduces source preference.
Studying end-to-end search with 12 agent models across three domains, the work finds each model prefers some sources and avoids others with broad agreement on which; an item satisfying one requirement fewer is selected about two-thirds of the time when it comes from a preferred source while the better item comes from a dispreferred source, but almost never in the reverse case; hiding source information weakens the preference and relabeling an item with a preferred source raises its selection rate; training that rewards better items can make a source a shortcut for requirement satisfaction, and missing information can trigger preconceptions about the source, while supplying missing information or a prompt countering those preconceptions reduces source preference.
Studying end-to-end search with 12 agent models across three domains, the work finds each model prefers some sources and avoids others with broad agreement on which; an item satisfying one requirement fewer is selected about two-thirds of the time when it comes from a preferred source while the better item comes from a dispreferred source, but almost never in the reverse case; hiding source information weakens the preference and relabeling an item with a preferred source raises its selection rate; training that rewards better items can make a source a shortcut for requirement satisfaction, and missing information can trigger preconceptions about the source, while supplying missing information or a prompt countering those preconceptions reduces source preference.