作者位次 #3/7ACL 2026Poster· Natural Language Processing
I m R easoner: Improving Memory-based Language Models for Reasoning-in-a-Haystack Tasks
Ching-Yun Ko · Payel Das · Sihui Dai · Georgios Kollias · Subhajit Chaudhury · Aurelie C. Lozano · Pin-Yu Chen
摘要
Reasoning over long contexts remains a major challenge for language models, particularly when solving tasks that require integrating multiple facts in sequence or generalizing to new distributions. We argue that this difficulty stems from a lack of structural inductive bias. Recently, alternative...