RLVR Datasets and Where to Find Them: Tracing Data Lineage for Better Training Data

• URL: http://arxiv.org/abs/2605.26971v1
• Date: Tue, 26 May 2026 12:57:12 GMT
• Title: RLVR Datasets and Where to Find Them: Tracing Data Lineage for Better Training Data
• Abstract summary: Atomic-source Tracing via Lineage-Aware Search (ATLAS) is a framework for tracing RLVR datasets back to their atomic sources.<n>Our analysis reveals that most RLVR datasets are variants of a small set of shared upstream sources, with few genuinely new data, and many facing data contamination risks.<n>We propose Source-level Counterfactual Attribution (SCA) as a guiding principle to curate a decontaminated training dataset with concentrated learning signals.
• Score: 15.904355053647913
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The proliferation of Reinforcement Learning from Verifiable Rewards (RLVR) datasets has exacerbated provenance collapse due to unclear lineage among existing datasets. To bridge this fragmented RLVR data landscape, we propose Atomic-source Tracing via Lineage-Aware Search (ATLAS), a systematic framework for tracing RLVR datasets back to their atomic sources, attributing over 99.7% of 1.45M instances to 20 atomic sources. Our analysis reveals that most RLVR datasets are variants of a small set of shared upstream sources, with few introducing genuinely new data, and many facing data contamination risks. These findings naturally motivate us to curate a new RLVR dataset, DAPO++, and to benchmark existing datasets from a lineage-aware perspective. To this end, we propose Source-level Counterfactual Attribution (SCA) as a guiding principle to curate a decontaminated training dataset with concentrated learning signals. Essentially, SCA measures a sample's marginal utility by comparing per-atomic-source RL checkpoints against a shared base model. Building upon these attribution signals, we further design a composite dataset quality score Q that strongly correlates with downstream RLVR performance. Experiments on Qwen3 series models verify that DAPO++ consistently improves performance on held-out benchmarks, while Q reliably predicts downstream RLVR training effectiveness. Our code and data is available at https://github.com/Celine-hxy/ATLAS.

Related papers

• On the Implicit Reward Overfitting and the Low-rank Dynamics in RLVR [51.935533482549545]
  We show that RLVR may exhibit implicit reward overfitting to the training dataset.<n>The model can achieve satisfactory performance on the test set even when its rewards remain relatively low during the training process.
  arXiv  Detail & Related papers  (2026-05-07T16:30:28Z)
• OpenDataArena: A Fair and Open Arena for Benchmarking Post-Training Dataset Value [74.80873109856563]
  OpenDataArena (ODA) is a holistic and open platform designed to benchmark the intrinsic value of post-training data.<n>ODA establishes a comprehensive ecosystem comprising four key pillars: (i) a unified training-evaluation pipeline that ensures fair, open comparisons across diverse models; (ii) a multi-dimensional scoring framework that profiles data quality along tens of distinct axes; and (iii) an interactive data lineage explorer to visualize dataset genealogy and dissect component sources.
  arXiv  Detail & Related papers  (2025-12-16T03:33:24Z)
• Masked-and-Reordered Self-Supervision for Reinforcement Learning from Verifiable Rewards [13.064343544668283]
  We propose MR-RLVR (Masked-and-Reordered RLVR), which constructs process-level self-supervised rewards via "masked-then-fill" and "step reordering"<n>We implement MR-RLVR on Qwen2.5-3B and DeepSeek-R1-Distill-Qwen-1.5B, and evaluate on AIME24, AIME25, AMC23, and MATH500.
  arXiv  Detail & Related papers  (2025-11-21T18:23:04Z)
• RL-Selector: Reinforcement Learning-Guided Data Selection via Redundancy Assessment [10.284993431741377]
  We introduce the concept of epsilon-sample cover, which quantifies sample redundancy based on inter-sample relationships.<n>We reformulate data selection as a reinforcement learning process and propose RL-Selector.<n>Our method consistently outperforms existing state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-06-26T06:28:56Z)
• Provably Efficient Online RLHF with One-Pass Reward Modeling [70.82499103200402]
  Reinforcement Learning from Human Feedback has shown remarkable success in aligning Large Language Models with human preferences.<n>Online RLHF has emerged as a promising direction, enabling iterative data collection and refinement.<n>We propose a one-pass reward modeling method that eliminates the need to store historical data and achieves constant-time updates per iteration.
  arXiv  Detail & Related papers  (2025-02-11T02:36:01Z)
• A Distribution-Aware Flow-Matching for Generating Unstructured Data for Few-Shot Reinforcement Learning [1.0709300917082865]
  We introduce a distribution-aware flow matching approach to generate synthetic unstructured data for few-shot reinforcement learning.<n>Our approach addresses key challenges in traditional model-based RL, such as overfitting and data correlation.<n>Results demonstrate that our method achieves stable convergence in terms of maximum Q-value while enhancing frame rates by 30% in the initial timestamps.
  arXiv  Detail & Related papers  (2024-09-21T15:50:59Z)
• Beyond Uniform Sampling: Offline Reinforcement Learning with Imbalanced Datasets [53.8218145723718]
  offline policy learning is aimed at learning decision-making policies using existing datasets of trajectories without collecting additional data. We argue that when a dataset is dominated by suboptimal trajectories, state-of-the-art offline RL algorithms do not substantially improve over the average return of trajectories in the dataset. We present a realization of the sampling strategy and an algorithm that can be used as a plug-and-play module in standard offline RL algorithms.
  arXiv  Detail & Related papers  (2023-10-06T17:58:14Z)
• Replication: Contrastive Learning and Data Augmentation in Traffic Classification Using a Flowpic Input Representation [47.95762911696397]
  We reproduce [16] on the same datasets and replicate its most salient aspect (the importance of data augmentation) on three additional public datasets. While we confirm most of the original results, we also found a 20% accuracy drop on some of the investigated scenarios due to a data shift in the original dataset.
  arXiv  Detail & Related papers  (2023-09-18T12:55:09Z)
• ORL-AUDITOR: Dataset Auditing in Offline Deep Reinforcement Learning [42.87245000172943]
  offline deep reinforcement learning ( offline DRL) is frequently used to train models on pre-collected datasets. We propose ORL-AUDITOR, which is the first trajectory-level dataset auditing mechanism for offline DRL scenarios. Our experiments on multiple offline DRL models and tasks reveal the efficacy of ORL-AUDITOR, with auditing accuracy over 95% and false positive rates less than 2.88%.
  arXiv  Detail & Related papers  (2023-09-06T15:28:43Z)
• D4RL: Datasets for Deep Data-Driven Reinforcement Learning [119.49182500071288]
  We introduce benchmarks specifically designed for the offline setting, guided by key properties of datasets relevant to real-world applications of offline RL. By moving beyond simple benchmark tasks and data collected by partially-trained RL agents, we reveal important and unappreciated deficiencies of existing algorithms.
  arXiv  Detail & Related papers  (2020-04-15T17:18:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.