ReAgg: Reliability-Aware Aggregation of Token Evidence for Machine-Generated Text Detection

The widespread use of large language models makes machine-generated text detection crucial for content provenance. Existing detectors typically aggregate token-level statistics into document scores via fixed rules, such as simple means or ratios. However, such uniform aggregation treats every token equally, regardless of its position in the text or how consistently the two reference models represent it. We derive an exact variance excess identity that quantifies the cost of weighting mismatch and support loss, and use this result to guide the construction of reliability weights from two observable cues. Specifically, we propose ReAgg, a reliability-aware aggregation framework that computes a contribution weight for each token using two computable cues: (i) its position in the text, and (ii) the representational stability between a base model and its instruction-tuned counterpart. It further adapts these weights to four scoring interfaces—means, sums, ratios, and spectral statistics—without training any additional model. We evaluate ReAgg across eight carriers spanning five benchmark families (18 subsets), improving macro-average direction-corrected AUROC from $75.22\%$ to $77.54\%$ ($+2.32$ pp) across the 13 primary subsets. Ablations support the empirical complementarity of the position and hidden-discrepancy cues.