English
Related papers

Related papers: Generating Pretraining Tokens from Organic Data fo…

200 papers

Large language models (LLMs) with extended context windows enable tasks requiring extensive information integration but are limited by the scarcity of high-quality, diverse datasets for long-context instruction tuning. Existing data…

Computation and Language · Computer Science 2025-02-25 Jiaxi Li , Xingxing Zhang , Xun Wang , Xiaolong Huang , Li Dong , Liang Wang , Si-Qing Chen , Wei Lu , Furu Wei

Token filtering has been proposed to enhance the utility of large language models (LLMs) by eliminating inconsequential tokens during training. While usingfewer tokens is expected to reduce computational workloads, existing methods have not…

Machine Learning · Computer Science 2026-03-20 Di Chai , Pengbo Li , Feiyuan Zhang , Yilun Jin , Han Tian , Kaiqiang Xu , Binhang Yuan , Dian Shen , Junxue Zhang , Kai Chen

The development of robust clinical decision support systems is frequently impeded by the scarcity of high-fidelity, privacy-preserving biomedical data. While Generative Large Language Models (LLMs) offer a promising avenue for synthetic…

Machine Learning · Computer Science 2026-04-03 Arshia Ilaty , Hossein Shirazi , Amir Rahmani , Hajar Homayouni

This paper introduces PRIMETIME, a synthetic generator that supports both benchmarking and fine-tuning of two primitive operations underlying temporal reasoning in Large Language Models (LLMs): parsing and arithmetic on datetimes. Existing…

Neural and Evolutionary Computing · Computer Science 2026-05-08 Edward Gaere , Florian Wangenheim

While synthetic data generation with large language models (LLMs) is widely used in post-training pipelines, existing approaches typically generate full outputs before applying quality filters, leading to substantial token waste on samples…

Artificial Intelligence · Computer Science 2026-05-15 Anjir Ahmed Chowdhury , Syed Zawad , Feng Yan

The application of machine learning on tabular data in specialized domains is severely limited by data scarcity. While generative models offer a solution, traditional methods falter in low-data regimes, and recent Large Language Models…

Machine Learning · Computer Science 2025-08-05 Siyi Liu , Yujia Zheng , Yongqi Zhang

Large language models (LLMs) have demonstrated strong capabilities in programming and mathematical reasoning tasks, but are constrained by limited high-quality training data. Synthetic data can be leveraged to enhance fine-tuning outcomes,…

Machine Learning · Computer Science 2025-04-28 Caia Costello , Simon Guo , Anna Goldie , Azalia Mirhoseini

Recently, synthetic data generation and realistic rendering has advanced tasks like target tracking and human pose estimation. Simulations for most robotics applications are obtained in (semi)static environments, with specific sensors and…

Computer Vision and Pattern Recognition · Computer Science 2023-05-29 Elia Bonetto , Chenghao Xu , Aamir Ahmad

Synthetic data offers a scalable solution for vision-language pre-training, yet current state-of-the-art methods typically rely on scaling up a single generative backbone, which introduces generator-specific spectral biases and limits…

Computer Vision and Pattern Recognition · Computer Science 2026-02-03 Leonardo Brusini , Cristian Sbrolli , Eugenio Lomurno , Toshihiko Yamasaki , Matteo Matteucci

Developing efficient CUDA kernels is a fundamental yet challenging task in the generative AI industry. Recent research leverages Large Language Models (LLMs) to automatically convert PyTorch reference implementations to CUDA kernels,…

Computation and Language · Computer Science 2026-05-28 Siqi Guo , Ming Lin , Tianbao Yang

Over recent years, an increasing amount of compute and data has been poured into training large language models (LLMs), usually by doing one-pass learning on as many tokens as possible randomly selected from large-scale web corpora. While…

Computation and Language · Computer Science 2023-08-24 Kushal Tirumala , Daniel Simig , Armen Aghajanyan , Ari S. Morcos

Training high-performing Small Language Models (SLMs) remains costly, even with knowledge distillation and pruning from larger teacher models. Existing work often faces three key challenges: (1) information loss from hard pruning, (2)…

Computation and Language · Computer Science 2025-12-19 Jitai Hao , Qiang Huang , Hao Liu , Xinyan Xiao , Zhaochun Ren , Jun Yu

Synthetic images are increasingly used to augment object-detection training sets, but reliably evaluating a synthetic dataset before training remains difficult: standard global generative metrics (e.g., FID) often do not predict downstream…

Computer Vision and Pattern Recognition · Computer Science 2026-02-24 Vasile Marian , Yong-Bin Kang , Alexander Buddery

LLMs are widely used in complex AI applications. These applications underscore the need for LLM outputs to adhere to a specific format, for their integration with other components in the systems. Typically the format rules e.g., for data…

Machine Learning · Computer Science 2024-11-07 Shubham Ugare , Tarun Suresh , Hangoo Kang , Sasa Misailovic , Gagandeep Singh

Recent English Common Crawl datasets like FineWeb-Edu and DCLM achieved significant benchmark gains via aggressive model-based filtering, but at the cost of removing 90% of data. This limits their suitability for long token horizon…

Computation and Language · Computer Science 2025-06-03 Dan Su , Kezhi Kong , Ying Lin , Joseph Jennings , Brandon Norick , Markus Kliegl , Mostofa Patwary , Mohammad Shoeybi , Bryan Catanzaro

Deploying LLMs raises two coupled challenges: (1) monitoring--estimating where a model underperforms as traffic and domains drift--and (2) improvement--prioritizing data acquisition to close the largest performance gaps. We test whether an…

Computation and Language · Computer Science 2026-05-27 Pedro Memoli Buffa , Luciano Del Corro

Deep generative models for tabular data (GANs, diffusion models, and LLM-based generators) exhibit highly non-uniform behavior across datasets; the best-performing synthesizer family depends strongly on distributional stressors such as…

Machine Learning · Computer Science 2026-04-02 Hochan Son , Xiaofeng Lin , Jason Ni , Guang Cheng

Large Language Models (LLMs) are increasingly used to generate synthetic textual data for training smaller specialized models. However, a comparison of various generation strategies for low-resource language settings is lacking. While…

Computation and Language · Computer Science 2025-09-22 Tatiana Anikina , Jan Cegin , Jakub Simko , Simon Ostermann

Mechanism-targeted synthetic data is increasingly proposed as a way to steer pretraining toward desirable capabilities, but it remains unclear how such interventions should be evaluated. We study this question for in-context learning (ICL)…

Computation and Language · Computer Science 2026-03-17 Mohammed Sabry , Anya Belz

Large language models (LLMs) excel in program synthesis, yet their capacity for neural architecture design -- balancing syntactic reliability, performance, and structural novelty -- remains underexplored. We present a closed-loop…

Machine Learning · Computer Science 2026-04-17 Waleed Khalid , Dmitry Ignatov , Radu Timofte