arXiv Machine Learning October 23, 2026 By Zhizhuo Kou, Sizai Hou, Long Chen, Jian Yang, Junyu Luo, Jingshu Peng, Wenkai Zhang, Sirui Han, Yike Guo

Inherited, Not Invented: Tracing the Provenance of Wait in Reasoning Models

Read the original on arXiv Machine Learning →

The paper investigates the origin of the word "wait" in reasoning models, arguing that it is inherited from training data rather than invented by reinforcement learning (RL). By analyzing training histories and using label‑free metrics, the authors show that RL improves accuracy without adding the marker, while the marker appears during mid‑training or fine‑tuning on reasoning traces. They also demonstrate that marker frequency alone does not prove RL‑induced reasoning and warn against relying on automated judges for evaluation.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

Jul 20

Understanding Reasoning from Pretraining to Post-Training

arXiv:2607. 16097v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become central to improving large language models (LLMs) on complex reasoning tasks, yet RL post-training is largely studied in isolation from the pretraining that precedes it.

By Jingyan Shen, Ang Li, Salman Rahman, Yifan Sun, Micah Goldblum, Matus Telgarsky, Pavel Izmailov

llms reinforcement-learning fine-tuning

arXiv AI

Aug 11

Understanding and Mitigating Premature Confidence for Better LLM Reasoning

arXiv:2605. 24396v2 Announce Type: replace Abstract: Long chains of thought (CoT) from current language models frequently contain logical gaps and unjustified leaps, limiting the gains from additional test-time compute.

By Jingchu Gai, Guanning Zeng, Christina Baek, Chen Wu, J. Zico Kolter, Andrej Risteski, Aditi Raghunathan

llms reinforcement-learning benchmarks safety

arXiv AI

Sep 1

Drop the Act: Probe-Filtered RL for Faithful Chain-of-Thought Reasoning

arXiv:2605.11467v2 Announce Type: replace-cross Abstract: Reasoning models post-hoc rationalize answers they have already committed to internally, producing chains of *reasoning theater*: deliberativ...

By Swapnil Parekh, Naman Goyal

llms reinforcement-learning benchmarks safety

arXiv Machine Learning

Jul 24

The Weight of Silence: A Causal Case for Weights Over the Scratchpad in Latent Chess Reasoning

arXiv:2607. 20952v1 Announce Type: new Abstract: Latent, or silent, reasoning lets language models carry out intermediate computation in continuous vector space instead of words, and is widely assumed to function as an internal scratchpad the model actively consults during inference.

By Ishan S. Kshirsagar

llms reinforcement-learning fine-tuning

arXiv Machine Learning

Oct 15

The Invisible Leash: Why RLVR May or May Not Escape Its Origin

arXiv:2507.14843v5 Announce Type: replace Abstract: Recent advances highlight Reinforcement Learning with Verifiable Rewards (RLVR) as a promising approach to enhancing LLM capabilities. However, it...

By Fang Wu, Weihao Xuan, Danny Xing, Ximing Lu, Mingjie Liu, Yi Dong, Zaid Harchaoui, Yejin Choi

llms reinforcement-learning benchmarks

arXiv AI

Jul 31

Probing the Origins of Reasoning Performance: Representational Quality for Mathematical Problem-Solving in RL vs. SFT Fine-Tuned Models

arXiv:2607. 26119v1 Announce Type: new Abstract: Large reasoning models trained via reinforcement learning (RL) have been increasingly shown to outperform their supervised fine-tuned (SFT) counterparts on mathematical reasoning tasks; Yet the mechanistic basis for this advantage remains unclear.

By Antyabha Rahman, Akshaj Gurugubelli, Omar Ankit, Kevin Zhu, Aishwarya Balwani

reinforcement-learning fine-tuning benchmarks

One email a morning, machine-written

Related stories

Understanding Reasoning from Pretraining to Post-Training

Understanding and Mitigating Premature Confidence for Better LLM Reasoning

Drop the Act: Probe-Filtered RL for Faithful Chain-of-Thought Reasoning

The Weight of Silence: A Causal Case for Weights Over the Scratchpad in Latent Chess Reasoning

The Invisible Leash: Why RLVR May or May Not Escape Its Origin

Probing the Origins of Reasoning Performance: Representational Quality for Mathematical Problem-Solving in RL vs. SFT Fine-Tuned Models