Intention Collapse: Intention-Level Metrics for Reasoning in Language Models

📝 Original Info

• Title: Intention Collapse: Intention-Level Metrics for Reasoning in Language Models
• ArXiv ID: 2601.01011
• Date: 2026-01-03
• Authors: Patricio Vera

📝 Abstract

Language generation maps a rich, high-dimensional internal state to a single token sequence. We study this many-to-one mapping through the lens of intention collapse: the projection from an internal intention space I to an external language space L. We introduce three cheap, model-agnostic metrics computed on a pre-collapse state I: (i) intention entropy Hint(I), (ii) effective dimensionality dim eff (I), and (iii) recoverability Recov(I), operationalized as probe AUROC for predicting eventual success. We evaluate these metrics in a 3 × 3 study across models (Mistral-7B, LLaMA-3.1-8B, Qwen-2.5-7B) and benchmarks (GSM8K, ARC-Challenge, AQUA-RAT), comparing baseline, chain-of-thought (CoT), and a babble control (n = 200 items per cell). CoT increases average accuracy from 34.2% to 47.3% (+13.1pp), driven by large gains on GSM8K but consistent degradations on ARC-Challenge. Across models, CoT induces distinct entropy regimes relative to baseline, ∆H = Hint(CoT) -Hint(Base): Mistral shows ∆H < 0 (lower-entropy CoT), whereas LLaMA shows ∆H > 0 (higher-entropy CoT), highlighting heterogeneity in CoT-induced internal uncertainty. Finally, probe AUROC is significantly above chance in a subset of settings and can dissociate from behavioral accuracy (e.g., high AUROC alon...

📄 Full Content

...(본문 내용이 길어 생략되었습니다. 사이트에서 전문을 확인해 주세요.)