← Back to all news
arXiv AI By Zilin Du, Bowen Yang, Boyang Albert Li

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

Read the original on arXiv AI →

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.

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 AI.

Related stories

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
llmsreinforcement-learningfine-tuning
More like this →
arXiv AI
Jun 4

Learning to Evaluate: Cost-Effective Model Evaluation on Unlabeled Data with Meta-Learning

arXiv:2605. 23595v2 Announce Type: replace-cross Abstract: The rapid advancement of machine learning has led to an unprecedented expansion of model ecosystems, making it increasingly difficult to assess the reliability of newly released models on unseen and unlabeled data.

By Trinh Pham, Viet Huynh, Hongzhi Yin, Quoc Viet Hung Nguyen, Thanh Tam Nguyen
fine-tuningbenchmarks
More like this →
arXiv Machine Learning
Jul 14

Metadata-Free Meta-Reweighted Direct Preference Optimization under Noisy Preference Labels

arXiv:2607. 09796v1 Announce Type: new Abstract: Direct Preference Optimization (DPO) has become an important method for aligning large language models (LLMs) with human preferences because it removes the need for explicit reward modeling and reinforcement learning optimization.

By Hua Qu, Yifan Li, Xiaodong Yuan
llmsnlpreinforcement-learningfine-tuningsafety
More like this →