SOURCE-LINKED INTELLIGENCE

Beyond Polarization: The Generative Constraint of Chain-of-Thought in Pointwise Reranking

arXiv · AI, language, vision and robotics · article · Aug 31, 2026 · UTC

In pointwise document reranking, Chain-of-Thought models typically underperform direct scoring models. While existing diagnostics attribute this to inferior classification, score polarization, or calibration breakdown, whether targeted training can bridge this gap remains unclear. Our empirical study first confirms that this gap is stable across scales up to 32B parameters, ruling out model and data capacity confounders. We then apply stress tests utilizing reinforcement learning, fine-grained supervision, and architectural decoupling to explicitly repair these deviations. Although these inter

Read original source ↗ Open in workspace

recordType

paper

region

Global