English
Related papers

Related papers: Beyond What to Select: A Plug-and-play Oscillatory…

200 papers

Instruction data selection aims to identify a high-quality subset from the training set that matches or exceeds the performance of the full dataset on target tasks. Existing methods focus on the instruction-to-response mapping, but neglect…

Machine Learning · Computer Science 2025-05-20 Wenya Guo , Zhengkun Zhang , Xumeng Liu , Ying Zhang , Ziyu Lu , Haoze Zhu , Xubo Liu , Ruxue Yan

Selecting high-quality training data from a larger pool is a crucial step when instruction-tuning language models, as carefully curated datasets often produce models that outperform those trained on much larger, noisier datasets. Automated…

Computation and Language · Computer Science 2025-06-23 Hamish Ivison , Muru Zhang , Faeze Brahman , Pang Wei Koh , Pradeep Dasigi

Instruction tuning has emerged as a critical paradigm for improving the capabilities and alignment of large language models (LLMs). However, existing iterative model-aware data selection methods incur significant computational overhead, as…

Machine Learning · Computer Science 2025-05-13 Xiaotian Lin , Yanlin Qi , Yizhang Zhu , Themis Palpanas , Chengliang Chai , Nan Tang , Yuyu Luo

As high-quality public text approaches exhaustion, a phenomenon known as the Data Wall, pre-training is shifting from more tokens to better tokens. However, existing methods either rely on heuristic static filters that ignore training…

Computation and Language · Computer Science 2026-02-10 Shaobo Wang , Xuan Ouyang , Tianyi Xu , Yuzheng Hu , Jialin Liu , Guo Chen , Tianyu Zhang , Junhao Zheng , Kexin Yang , Xingzhang Ren , Dayiheng Liu , Linfeng Zhang

Reinforcement learning with verifiable rewards (RLVR) has emerged as the leading approach for enhancing reasoning capabilities in large language models. However, it faces a fundamental compute and memory asymmetry: rollout generation is…

Machine Learning · Computer Science 2026-04-23 Yixuan Even Xu , Yash Savani , Fei Fang , J. Zico Kolter

Dynamic data selection aims to accelerate training with lossless performance. However, reducing training data inherently limits data diversity, potentially hindering generalization. While data augmentation is widely used to enhance…

Machine Learning · Computer Science 2025-05-13 Suorong Yang , Peng Ye , Furao Shen , Dongzhan Zhou

Selecting appropriate training data is crucial for effective instruction fine-tuning of large language models (LLMs), which aims to (1) elicit strong capabilities, and (2) achieve balanced performance across a diverse range of tasks.…

Computation and Language · Computer Science 2025-01-22 Qirun Dai , Dylan Zhang , Jiaqi W. Ma , Hao Peng

We introduce Goldilocks Selection, a technique for faster model training which selects a sequence of training points that are "just right". We propose an information-theoretic acquisition function -- the reducible validation loss -- and…

Data selection is designed to accelerate learning with preserved performance. To achieve this, a fundamental thought is to identify informative data samples with significant contributions to the training. In this work, we propose…

Machine Learning · Computer Science 2025-09-30 Ziheng Cheng , Zhong Li , Jiang Bian

Large-scale multimodal pre-trained models like CLIP rely heavily on high-quality training data, yet raw web-crawled datasets are often noisy, misaligned, and redundant, leading to inefficient training and suboptimal generalization. Existing…

Machine Learning · Computer Science 2026-02-06 Guanjie Cheng , Boyi Li , Lingyu Sun , Mengying Zhu , Yangyang Wu , Xinkui Zhao , Shuiguang Deng

Instruction tuning has become the de facto method to equip large language models (LLMs) with the ability of following user instructions. Usually, hundreds of thousands or millions of instruction-following pairs are employed to fine-tune the…

Computation and Language · Computer Science 2023-11-28 Qianlong Du , Chengqing Zong , Jiajun Zhang

Dynamic Data selection aims to accelerate training by prioritizing informative samples during online training. However, existing methods typically rely on task-specific handcrafted metrics or static/snapshot-based criteria to estimate…

Machine Learning · Computer Science 2026-05-14 Suorong Yang , Fangjian Su , Hai Gan , Ziqi Ye , Jie Li , Baile Xu , Furao Shen , Soujanya Poria

Recently, preference optimization methods such as DPO have significantly enhanced large language models (LLMs) in wide tasks including dialogue and question-answering. However, current methods fail to account for the varying difficulty…

Computation and Language · Computer Science 2024-12-31 Jingyuan Ma , Rui Li , Zheng Li , Lei Sha , Zhifang Sui

Pretraining data selection has the potential to improve language model pretraining efficiency by utilizing higher-quality data from massive web data corpora. Current data selection methods, which rely on either hand-crafted rules or larger…

Computation and Language · Computer Science 2024-11-19 Zichun Yu , Spandan Das , Chenyan Xiong

Data selection methods address a critical challenge in LLM post-training: effectively leveraging scarce, high-fidelity target data alongside abundant but imperfectly aligned general training data. In this work, we move beyond the…

Machine Learning · Computer Science 2026-05-11 Pingbang Hu , Xueshen Liu , Z. Morley Mao , Jiaqi W. Ma

Direct Preference Optimization (DPO) and its variants have become the de facto standards for aligning large language models (LLMs) with human preferences or specific goals. However, DPO requires high-quality preference data and suffers from…

Machine Learning · Computer Science 2024-11-12 Zhuotong Chen , Fang Liu , Jennifer Zhu , Wanyu Du , Yanjun Qi

To acquire a new skill, humans learn better and faster if a tutor, based on their current knowledge level, informs them of how much attention they should pay to particular content or practice problems. Similarly, a machine learning model…

Machine Learning · Computer Science 2021-06-18 Xinyi Wang , Hieu Pham , Paul Michel , Antonios Anastasopoulos , Jaime Carbonell , Graham Neubig

We introduce Predictive Batch Scheduling (PBS), a novel training optimization technique that accelerates language model convergence by dynamically prioritizing high-loss samples during batch construction. Unlike curriculum learning…

Artificial Intelligence · Computer Science 2026-02-20 Sumedh Rasal

Instruction-tuning language models has become a crucial step in aligning them for general use. Typically, this process involves extensive training on large datasets, incurring high training costs. In this paper, we introduce a novel…

Computation and Language · Computer Science 2024-02-19 Dheeraj Mekala , Alex Nguyen , Jingbo Shang

This work investigates the selection of high-quality pre-training data from massive corpora to enhance LMs' capabilities for downstream usage. We formulate data selection as a generalized Optimal Control problem, which can be solved…

Computation and Language · Computer Science 2025-03-20 Yuxian Gu , Li Dong , Hongning Wang , Yaru Hao , Qingxiu Dong , Furu Wei , Minlie Huang
‹ Prev 1 2 3 10 Next ›