Position Bias in Long-Context Retrieval: A Replication Study
Several studies report that language models use information in the middle of a long context less reliably than information at the start or end. We replicate these findings across eleven models and four context lengths using a controlled retrieval task. The effect is reproducible but smaller than originally reported, and largely disappears when the task format is held fixed across lengths. We release the harness and the full per-position results.