paper-with-me

홈 › Papers

The Unlearnability Phenomenon in RLVR for Language Models

2026-05-16 · Yulin Chen, He He, Chen Zhao arxiv

Reinforcement Learning with Verifiable Reward (RLVR) has proven effective in improving Large Language Model's (LLM) reasoning ability. However, the learning dynamics of RLVR remain underexplored. In this paper, we reveal a counterintuitive phenomenon: among hard examples that the model initially struggles with, a substantial subset remains unlearnable even when correct rollouts are present. To understand the phenomenon, we first demonstrate that existing optimization and sampling techniques fail to resolve unlearnability. With cross-example gradient analysis, we show that unlearnable examples have fundamental representation issue, characterized by low gradient similarity with the rest of the examples and ungeneralizable reasoning patterns. We further show that representation flaws are difficult to mitigate in RL, as data augmentation does not improve gradient similarity. Our study provides the first systematic characterization of unlearnable data in RLVR training and reveals fundamental limitations in current RL approaches for reasoning tasks. Code and data are available at \url{https://github.com/yulinchen99/unlearnability-rlvr}.

📄 PDF Abstract BibTeX arXiv:2605.16787

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningData Augmentation

Similar Papers 제목 키워드 기반

How Far Are We from True Unlearnability?

2025-09-09 · Kai Ye, Liangcai Su, Chenxiong Qian arxiv

High-quality data plays an indispensable role in the era of large models, but the use of unauthorized data for model training greatly damages the interests of data owners. To overcome this threat, several unlearnable met…

Representation LearningSemantic Segmentation

Reinforcement Learning for Reasoning in Large Language Models with One Training Example

2025-04-29 · Yiping Wang, Qing Yang, Zhiyuan Zeng, Liliang Ren 외

We show that reinforcement learning with verifiable reward using one training example (1-shot RLVR) is effective in incentivizing the math reasoning capabilities of large language models (LLMs). Applying RLVR to the base…

Domain GeneralizationMathMathematical Reasoning

SoK: Unlearnability and Unlearning for Model Dememorization

2026-05-12 · Mengying Zhang, Derui Wang, Ruoxi Sun, Xiaoyu Xia 외 arxiv

Advanced model dememorization methods, including availability poisoning (unlearnability) and machine unlearning, are emerging as key safeguards against data misuse in machine learning (ML). At the training stage, unlearn…

The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models

2025-10-02 · Phuc Minh Nguyen, Chinh D. La, Duy M. H. Nguyen, Nitesh V. Chawla 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a key method for improving Large Language Models' reasoning capabilities, yet recent evidence suggests it may paradoxically shrink the reasoning bounda…

Reinforcement LearningMathematical Reasoning

How You Begin is How You Reason: Driving Exploration in RLVR via Prefix-Tuned Priors

2026-05-09 · Yifan Xu, Junren Chen, Yifan Chen arxiv

Reinforcement learning with verifiable rewards (RLVR) recently thrives in large language model (LLM) reasoning tasks. However, the reward sparsity and the long reasoning horizon make effective exploration challenging. In…

Reinforcement Learning