English
Related papers

Related papers: The Delta Learning Hypothesis: Preference Tuning o…

200 papers

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

Domain adaptation of neural networks commonly relies on three training phases: pretraining, selected data training and then fine tuning. Data selection improves target domain generalization by training further on pretraining data identified…

Computation and Language · Computer Science 2021-09-17 Dan Iter , David Grangier

Preference alignment in Large Language Models (LLMs) has significantly improved their ability to adhere to human instructions and intentions. However, existing direct alignment algorithms primarily focus on relative preferences and often…

Machine Learning · Computer Science 2025-05-13 Shenao Zhang , Zhihan Liu , Boyi Liu , Yufeng Zhang , Yingxiang Yang , Yongfei Liu , Liyu Chen , Tao Sun , Zhaoran Wang

Aligning language models with human preferences relies on pairwise preference datasets. While some studies suggest that on-policy data consistently outperforms off -policy data for preference learning, others indicate that the advantages of…

Computation and Language · Computer Science 2025-05-06 Tianjian Li , Daniel Khashabi

Neural models often exploit superficial features to achieve good performance, rather than deriving more general features. Overcoming this tendency is a central challenge in areas such as representation learning and ML fairness. Recent work…

Computation and Language · Computer Science 2020-10-12 Rohan Jha , Charles Lovering , Ellie Pavlick

The rise of large language models (LLMs) has created a significant disparity: industrial research labs with their computational resources, expert teams, and advanced infrastructures, can effectively fine-tune LLMs, while individual…

Language model pretraining involves training on extensive corpora, where data quality plays a pivotal role. In this work, we aim to directly estimate the contribution of data during pretraining and select pretraining data in an efficient…

Computation and Language · Computer Science 2025-08-05 Kashun Shum , Yuzhen Huang , Hongjian Zou , Qi Ding , Yixuan Liao , Xiaoxin Chen , Qian Liu , Junxian He

Most large language models are fine-tuned using either expensive human-annotated data or GPT-4 generated data which cannot guarantee performance in certain domains. We argue that although the web-crawled data often has formatting errors…

Computation and Language · Computer Science 2024-08-16 Jing Zhou , Chenglin Jiang , Wei Shen , Xiao Zhou , Xiaonan He

How can we train models whose post-trained capabilities survive subsequent fine-tuning? Rather than focusing on downstream interventions to mitigate forgetting of upstream capabilities, we study how upstream training choices - that is, the…

Machine Learning · Computer Science 2026-05-14 Lawrence Feng , Gaurav R. Ghosal , Jacob Mitchell Springer , Ziqian Zhong , Aditi Raghunathan

Weakly supervised learning is a popular approach for training machine learning models in low-resource settings. Instead of requesting high-quality yet costly human annotations, it allows training models with noisy annotations obtained from…

Computation and Language · Computer Science 2023-09-19 Dawei Zhu , Xiaoyu Shen , Marius Mosbach , Andreas Stephan , Dietrich Klakow

Large language models are first pre-trained on trillions of tokens and then instruction-tuned or aligned to specific preferences. While pre-training remains out of reach for most researchers due to the compute required, fine-tuning has…

Computation and Language · Computer Science 2024-06-10 Megh Thakkar , Quentin Fournier , Matthew D Riemer , Pin-Yu Chen , Amal Zouaq , Payel Das , Sarath Chandar

Aligning the output of Large Language Models (LLMs) with human preferences (e.g., by means of reinforcement learning with human feedback, or RLHF) is essential for ensuring their effectiveness in real-world scenarios. Despite significant…

Artificial Intelligence · Computer Science 2024-10-23 Pietro Bernardelle , Gianluca Demartini

Deep Neural Networks are well known for efficiently fitting training data, yet experiencing poor generalization capabilities whenever some kind of bias dominates over the actual task labels, resulting in models learning "shortcuts". In…

Machine Learning · Computer Science 2024-08-12 Pietro Morerio , Ruggero Ragonesi , Vittorio Murino

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

Aligning language models with human preferences through reinforcement learning from human feedback is crucial for their safe and effective deployment. The human preference is typically represented through comparison where one response is…

Machine Learning · Computer Science 2025-07-15 Hoang Anh Just , Ming Jin , Anit Sahu , Huy Phan , Ruoxi Jia

In the Mixup training paradigm, a model is trained using convex combinations of data points and their associated labels. Despite seeing very few true data points during training, models trained using Mixup seem to still minimize the…

Machine Learning · Computer Science 2022-02-22 Muthu Chidambaram , Xiang Wang , Yuzheng Hu , Chenwei Wu , Rong Ge

In the last decade, motivated by the success of Deep Learning, the scientific community proposed several approaches to make the learning procedure of Neural Networks more effective. When focussing on the way in which the training data are…

Machine Learning · Computer Science 2021-06-22 Simone Marullo , Matteo Tiezzi , Marco Gori , Stefano Melacci

How can "weak teacher models" such as average human annotators or existing AI systems, effectively supervise LLMs to improve performance on hard reasoning tasks, especially those that challenge and requires expertise or daily practice from…

Machine Learning · Computer Science 2025-02-26 Xuan He , Da Yin , Nanyun Peng

Current instruction-tuned language models are exclusively trained with textual preference data and thus are often not aligned with the unique requirements of other modalities, such as speech. To better align language models with the speech…

The quality of an induced model by a learning algorithm is dependent on the quality of the training data and the hyper-parameters supplied to the learning algorithm. Prior work has shown that improving the quality of the training data…

Machine Learning · Statistics 2014-03-14 Michael R. Smith , Tony Martinez , Christophe Giraud-Carrier