English
Related papers

Related papers: Hard Examples Are All You Need: Maximizing GRPO Po…

200 papers

Reinforcement learning has emerged as a powerful paradigm for unlocking reasoning capabilities in language models. However, relying on sparse rewards makes this process highly sample-inefficient, as models must navigate vast search spaces…

Machine Learning · Computer Science 2026-05-11 Ilia Mahrooghi , Aryo Lotfi , Emmanuel Abbe

Modern code generation models exhibit longer outputs, accelerated capability growth, and changed training dynamics, rendering traditional training methodologies, algorithms, and datasets ineffective for improving their performance. To…

Machine Learning · Computer Science 2026-03-10 Zongqian Li , Shaohan Huang , Zewen Chi , Yixuan Su , Lexin Zhou , Li Dong , Nigel Collier , Furu Wei

The NLP community has long advocated for the construction of multi-annotator datasets to better capture the nuances of language interpretation, subjectivity, and ambiguity. This paper conducts a retrospective study to show how performance…

Computation and Language · Computer Science 2023-10-24 Pritam Kadasi , Mayank Singh

Instruct models, obtained from various instruction tuning or post-training steps, are commonly deemed superior and more usable than their base counterpart. While the model gains instruction following ability, instruction tuning may lead to…

Computation and Language · Computer Science 2025-09-23 Ozan İrsoy , Pengxiang Cheng , Jennifer L. Chen , Daniel Preoţiuc-Pietro , Shiyue Zhang , Duccio Pappadopulo

Modern NLP systems require high-quality annotated data. In specialized domains, expert annotations may be prohibitively expensive. An alternative is to rely on crowdsourcing to reduce costs at the risk of introducing noise. In this paper we…

Computation and Language · Computer Science 2019-05-21 Yinfei Yang , Oshin Agarwal , Chris Tar , Byron C. Wallace , Ani Nenkova

Large language models trained with reinforcement learning with verifiable rewards tend to trade accuracy for length--inflating response lengths to achieve gains in accuracy. While longer answers may be warranted for harder problems, many…

Computation and Language · Computer Science 2025-08-14 Vaishnavi Shrivastava , Ahmed Awadallah , Vidhisha Balachandran , Shivam Garg , Harkirat Behl , Dimitris Papailiopoulos

The performance of fine-tuning pre-trained language models largely depends on the hyperparameter configuration. In this paper, we investigate the performance of modern hyperparameter optimization methods (HPO) on fine-tuning pre-trained…

Computation and Language · Computer Science 2021-06-18 Xueqing Liu , Chi Wang

Post-training has become effective for high-level generation, but its role in low-level vision remains underexplored. Existing image restoration methods often rely on fixed pixel-wise fitting to ground-truth images, which can lead to…

Computer Vision and Pattern Recognition · Computer Science 2026-05-28 Haoxuan Xu , Yi Liu , Tianfu Li , Ruolin Shen , Boyuan Jiang , Jinlong Peng , Donghao Luo , Xiaobin Hu , Shuicheng Yan , Haoang Li

Current foundation models have shown impressive performance across various tasks. However, several studies have revealed that these models are not effective for everyone due to the imbalanced geographical and economic representation of the…

Computer Vision and Pattern Recognition · Computer Science 2024-03-13 Oana Ignat , Longju Bai , Joan Nwatu , Rada Mihalcea

Post-training (via supervised fine-tuning) improves instruction-following, but often induces semantic mode collapse by biasing models toward low-entropy fine-tuning data at the expense of the high-entropy pretraining distribution.…

Active learning (AL) is a widely-used training strategy for maximizing predictive performance subject to a fixed annotation budget. In AL one iteratively selects training examples for annotation, often those for which the current model is…

Machine Learning · Computer Science 2019-11-05 David Lowell , Zachary C. Lipton , Byron C. Wallace

Group Relative Policy Optimization (GRPO) has become a standard approach for training mathematical reasoning models; however, its reliance on multiple completions per prompt makes training computationally expensive. Although recent work has…

Machine Learning · Computer Science 2026-01-15 Kangda Wei , Ruihong Huang

Text-to-Image (T2I) generation has achieved remarkable progress in recent years. Meanwhile, reinforcement learning methods, particularly those based on Group Relative Policy Optimization (GRPO), have attracted widespread attention and been…

Computer Vision and Pattern Recognition · Computer Science 2026-05-19 Baoteng Li , Xianghao Zang , Xinran Wang , Xiangyu Na , Zhixiang He , Hao Sun , Chi Zhang , Zhongjiang He , Tianwei Cao , Kongming Liang , Zhanyu Ma

Many machine learning systems today are trained on large amounts of human-annotated data. Data annotation tasks that require a high level of competency make data acquisition expensive, while the resulting labels are often subjective,…

Machine Learning · Computer Science 2020-04-08 Emmanouil Antonios Platanios , Maruan Al-Shedivat , Eric Xing , Tom Mitchell

Group Relative Policy Optimization (GRPO) has become the dominant method for reinforcement learning with verifiable rewards in large language models, but it suffers from two critical limitations: gradient vanishing and diversity collapse.…

Machine Learning · Computer Science 2026-05-20 Khiem Le , Phuc Nguyen , Youssef Mroueh , Chi-Heng Lin , Shangqian Gao , Ting Hua , Nitesh V. Chawla

This paper introduces Completion Pruning Policy Optimization (CPPO) to accelerate the training of reasoning models based on Group Relative Policy Optimization (GRPO). GRPO, while effective, incurs high training costs due to the need to…

Artificial Intelligence · Computer Science 2025-11-11 Zhihang Lin , Mingbao Lin , Yuan Xie , Rongrong Ji

We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and evaluation. Existing research is mixed regarding whether training on easier or harder data…

Computation and Language · Computer Science 2025-11-27 Yeganeh Kordi , Nihal V. Nayak , Max Zuo , Ilana Nguyen , Stephen H. Bach

We explore the utilities of explicit negative examples in training neural language models. Negative examples here are incorrect words in a sentence, such as "barks" in "*The dogs barks". Neural language models are commonly trained only on…

Computation and Language · Computer Science 2020-06-09 Hiroshi Noji , Hiroya Takamura

We classify and re-examine some of the current approaches to improve the performance-computes trade-off of language models, including (1) non-causal models (such as masked language models), (2) extension of batch length with efficient…

Computation and Language · Computer Science 2020-09-16 Aran Komatsuzaki

When tasked with supporting multiple languages for a given problem, two approaches have arisen: training a model for each language with the annotation budget divided equally among them, and training on a high-resource language followed by…

Computation and Language · Computer Science 2022-04-05 Joel Ruben Antony Moniz , Barun Patra , Matthew R. Gormley