这篇论文发现DeepSeek-R1推理非收敛在早期就能从内部表征检测到，未来可做自适应计算分配，研究推理效率的朋友可以看看。
DeepSeek-R1-Distill-Qwen-7B等思维链模型在推理时呈现双峰收敛模式：在令牌预算内结束（收敛）或耗尽预算无结论（非收敛）。实验显示，收敛的推理在AIME 1983-2024上准确率达90.3%，非收敛仅6.6%，总体收敛率62.0%。通过训练线性探针在隐藏状态激活（token 50-300）上，层20在token 150处达到AUC 0.608（±0.080，五折交叉验证）。激活探针优于基于令牌熵和重复统计的行为基线。扫描级置换检验p=0.063（100,000次置换），表明样本量不足以在常规阈值下确认信号，但收敛命运在生成结束前部分编码于中间表征。
Token Budget Saturation and Mechanistic Early Detection of Reasoning Non-Convergence in Chain-of-Thought Models
Chain-of-thought reasoning models such as DeepSeek-R1-Distill-Qwen-7B exhibit a bimodal convergence pattern: generations either terminate within a token budget (converged) or exhaust it without reaching a conclusion (non-converged). We characterize this phenomenon empirically, showing that converged generations achieve 90.3% accuracy on AIME 1983-2024 while non-converged ones achieve only 6.6%, with an overall convergence rate of 62.0%. We then ask whether this outcome is detectable early in the thinking chain using internal model representations. Training linear probes on hidden-state activations at token positions 50-300, we find that layer-20 activations at token 150 achieve AUC 0.608 (+-0.080, 5-fold CV), reliably above chance even at token 50. Activation probes consistently outperform behavioral baselines derived from token entropy and repetition statistics. A sweep-level permutation test yields p=0.063 (100,000 permutations), consistent with a modest signal that our sample size cannot confirm at conventional thresholds. These findings suggest that convergence fate is partially encoded in intermediate representations well before the generation ends, opening a path toward early-exit inference and adaptive compute allocation.