כתבה
arXiv cs.CL ·
מילים דוברות יותר מסדר: בירור התנהגותי של Gemma 4
Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4
גמה 4 נוטה להעדיף את תיאור המקור על פני סדר הקריאה
תקציר מקורי באנגליתarXiv:2609.30716v1 Announce Type: new Abstract: When a language model receives two conflicting documents as input, how does it decide which one to prioritize? Does it rely on how the sources are framed or the presentation order of the documents? We evaluated this behavior on Google's pre-trained Gemma 4-e4b model across a targeted behavioral suite (n = 13 items, 784 forward passes in short, single-turn contexts) using a completely counterbalanced experimental design. This setup allowed us to mathematically isolate the specific effects of source framing and reading position, while ensuring the model's natural vocabulary biases were canceled out. Across ten test conditions, we discovered the following: 1. Source framing heavily overpowers reading position. When directly competing, the semant
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית