arXiv AI

Dr. Post-Training: A Data Regularization Perspective on LLM Post-Training

The paper introduces Dr. Post-Training, a framework that treats general training data as a data‑induced regularizer to prevent overfitting to scarce target objectives, rather than merely a pool for selection. It constructs a feasible set of model update directions from general data and projects target‑driven updates onto this set, unifying standard training and existing data‑selection methods as special cases along a bias‑variance spectrum. Experiments on SFT, RLHF, and RLVR demonstrate that these methods outperform both standard training and state‑of‑the‑art data‑selection baselines, especially when general and target data are highly misaligned, while system optimizations keep overhead minimal.

arXiv Machine Learning

Aug 19

Data-DPO: Direct Preference Optimization for Target Model Data Selection in LLM Post-Training

Data-DPO is a target model‑oriented supervised fine‑tuning data selection method that uses one‑step probing of the target model to generate pairwise data preferences, trains a lightweight reward model to capture these preferences, and then selects a training subset by combining target‑model preference, external quality scores, and marginal diversity. Experiments on Vision‑Flan and LLaVA‑CoT demonstrate that Data‑DPO consistently outperforms existing data selection baselines across multiple data budgets and even surpasses full data training performance.

By Peng Sun, Yi Yang, Antong Zhang, Chunxiao Li, Yanbo Wang, Dianbo Liu, xin chen, Kai Yu, Lu Chen, Tianfan Fu

llms reinforcement-learning fine-tuning

arXiv AI

Oct 21

The Shape of the DPO Hyperparameter Landscape: What to Measure, How It Moves with Scale, and When Not to Transfer It

arXiv:2610.27895v1 Announce Type: cross Abstract: Direct preference optimisation (DPO) is widely used for language model post-training, but its performance depends strongly on the choice of its hyper...

By Hannan Mahadik, David Salinas, Jacek Golebiowski, Aaron Klein

llms reinforcement-learning benchmarks

arXiv Machine Learning

Jun 18

BLADE: Scalable Bi-level Adaptive Data Selection for LLM Training

arXiv:2606. 18650v1 Announce Type: new Abstract: As Large Language Model (LLM) datasets scale to trillions of tokens, data selection has emerged as a critical frontier to filter out uninformative noise and construct adaptive learning trajectories.

By Jiaxing Wang, Deping Xiang, Jin Xu, Zirui Liu, Zicheng Zhang, Guoqiang Gong, Jun Fang, Chao Liu, Pengzhang Liu, Tongxuan Liu, Ke Zhang, Qixia Jiang

llms benchmarks

arXiv AI

Oct 2

Training-Aware Target Coverage for Synthetic Data Selection

arXiv:2610.00814v1 Announce Type: cross Abstract: Synthetic data are increasingly used to scale LLM training, yet more synthetic data do not necessarily produce better models. Useful synthetic data m...

By Yang Ba, Michelle V. Mancenido, Rong Pan

llms fine-tuning

arXiv Machine Learning

Sep 30

UnlearningSoup: Is Repeated Tuning Necessary for Large Language Model Unlearning?

arXiv:2609.37076v1 Announce Type: new Abstract: Large language models trained on vast corpora inherently risk memorizing harmful content that may later re-emerge in their outputs. To mitigate this is...

By Puning Yang, Qizhou Wang, Junchi Yu, Bo Han, Xiuying Chen

llms

arXiv AI

Jun 16

Utility-Diversity Aware Online Batch Selection for LLM Supervised Fine-tuning

arXiv:2510. 16882v4 Announce Type: replace-cross Abstract: Supervised fine-tuning (SFT) is a commonly used technique to adapt large language models (LLMs) to downstream tasks.

By Heming Zou, Yixiu Mao, Yun Qu, Qi Wang, Xiangyang Ji

llms rag fine-tuning benchmarks safety

arXiv Computation and Language

Aug 21

EvoSelect: Data-Efficient LLM Evolution for Targeted Task Adaptation

arXiv:2604. 26170v2 Announce Type: replace Abstract: Adapting large language models (LLMs) to a targeted task efficiently and effectively remains a fundamental challenge.

By Ting-Wei Li, Sirui Chen, Jiaru Zou, Yingbing Huang, Tianxin Wei, Jingrui He, Hanghang Tong

llms benchmarks safety

arXiv AI

Sep 30

Loss-Guided Pretraining Data Selection for Time-Series Foundation Models

arXiv:2609.37255v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) are pretrained on heterogeneous collections containing billions of observations, yet their training windows are...

By Yike Li, Shaoxu Song, Jianmin Wang

arXiv AI

3d ago

Advancement or Acceleration? How Data Curation Improves LLM Training

arXiv:2610. 36983v1 Announce Type: cross Abstract: While data curation is widely credited with improving large language model training, a basic question remains unresolved: does curation expand what a model can ultimately learn, or merely reach the same solution faster?

By Jiaxing Wang, Zicheng Zhang, Deping Xiang, Jin Xu, Yong Li, Chao Liu, Pengzhang Liu, Qixia Jiang

llms

arXiv AI

Jun 15

Beyond LoRA: Is Sparsity-Induced Adaptation Better?

arXiv:2606. 13767v1 Announce Type: cross Abstract: Low-rank adaptation (LoRA) and its variants provide a memory- and compute-efficient alternative to full fine-tuning of pre-trained models.

By Elijah Cadenhead, Cristian McGee, Xin Li, El Houcine Bergou, Aritra Dutta

fine-tuning efficiency

arXiv AI

4d ago

Automatic Configuration of LLM Post-Training Pipelines

arXiv:2603.18773v2 Announce Type: replace-cross Abstract: LLM post-training pipelines that combine supervised fine-tuning and reinforcement learning are difficult to configure under realistic compute...

By Channe Chwa, Xinle Wu, Yao Lu

llms reinforcement-learning fine-tuning

arXiv AI

Oct 2

Scalable, Transferable Meta-network for Data Selection Requires a Different Loss (and Why the Obvious Choice is Problematic)

The paper introduces TESS, a scalable data‑selection framework that replaces per‑sample weights with a selection network to improve transferability across datasets and model sizes. It identifies instability in existing meta‑learning for training‑data selection (MTS) due to weight suppression and overreliance on easy features, and proposes a Pointwise Value Matching objective to address these issues. Experiments on large language model safety and instruction tuning show strong transfer from subsets to full corpora and from smaller to larger models.

By Zilin Du, Bowen Yang, Boyang Albert Li

llms fine-tuning safety

Related stories