SOURCE-LINKED INTELLIGENCE

Beyond Depth Truncation: Controlled Evaluation of Depth Utilization in Recursive Language Models

arXiv · AI, language, vision and robotics · article · Sep 17, 2026 · UTC

Depth-recurrent language models iteratively apply a small layer stack, decoupling per-token compute from distinct parameter count. To determine whether such a model genuinely utilizes its depth, both recurrence and layer-pruning literatures rely on a shared evaluation: truncating depth at inference time, plotting quality against retained depth fraction, and reading off the slope. While cheap and training-free, this metric suffers from an unexamined flaw: it extracts a single scalar from an intervention that alters multiple model properties simultaneously. Depth truncation concurrently reduces

Read original source ↗ Open in workspace

recordType

paper

region

Global