English
Related papers

Related papers: Iterative Finetuning is Mostly Idempotent

200 papers

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

Self-supervised representation learning often uses data augmentations to induce some invariance to "style" attributes of the data. However, with downstream tasks generally unknown at training time, it is difficult to deduce a priori which…

Large language models (LLMs) have shown remarkable abilities in diverse natural language processing (NLP) tasks. The LLMs generally undergo supervised fine-tuning (SFT) followed by preference alignment to be usable in downstream…

Computation and Language · Computer Science 2024-06-27 Shiva Kumar Pentyala , Zhichao Wang , Bin Bi , Kiran Ramnath , Xiang-Bo Mao , Regunathan Radhakrishnan , Sitaram Asur , Na , Cheng

Augmenting large language models (LLMs) with external tools is a promising approach to enhance their capabilities, especially for complex tasks. Synthesizing tool-use data through real-world simulations is an effective way to achieve this.…

Computation and Language · Computer Science 2025-11-10 Yirong Zeng , Xiao Ding , Yuxian Wang , Weiwen Liu , Wu Ning , Yutai Hou , Xu Huang , Duyu Tang , Dandan Tu , Bing Qin , Ting Liu

Data augmentation plays a pivotal role in enhancing and diversifying training data. Nonetheless, consistently improving model performance in varied learning scenarios, especially those with inherent data biases, remains challenging. To…

Machine Learning · Computer Science 2024-06-04 Xiaoling Zhou , Wei Ye , Zhemg Lee , Rui Xie , Shikun Zhang

Foundation models trained on web-scale data have revolutionized robotics, but their application to low-level control remains largely limited to behavioral cloning. Drawing inspiration from the success of the reinforcement learning stage in…

Machine Learning · Computer Science 2025-09-19 Seyed Kamyar Seyed Ghasemipour , Ayzaan Wahid , Jonathan Tompson , Pannag Sanketi , Igor Mordatch

Existing work has shown that o1-level performance can be achieved with limited data distillation, but most existing methods focus on unidirectional supervised fine-tuning (SFT), overlooking the intricate interplay between diverse reasoning…

Machine Learning · Computer Science 2025-09-17 Mengyi Deng , Xin Li , Tingyu Zhu , Zhicheng Yang , Zhijiang Guo , Wei Wang

The rapid progress in generative models has resulted in impressive leaps in generation quality, blurring the lines between synthetic and real data. Web-scale datasets are now prone to the inevitable contamination by synthetic data, directly…

Computer Vision and Pattern Recognition · Computer Science 2024-07-16 Damien Ferbach , Quentin Bertrand , Avishek Joey Bose , Gauthier Gidel

Existing LLMs-post-training techniques are broadly categorized into supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT). Each paradigm presents a distinct trade-off: (1) SFT excels at mimicking demonstration data, but can lead…

Machine Learning · Computer Science 2026-05-18 Zeyu Huang , Tianhao Cheng , Zihan Qiu , Zili Wang , Yinghui Xu , Edoardo M. Ponti , Ivan Titov

Language model (LM) post-training relies on two stages of human supervision: task demonstrations for supervised finetuning (SFT), followed by preference comparisons for reinforcement learning from human feedback (RLHF). As LMs become more…

Machine Learning · Computer Science 2025-01-15 Yaowen Ye , Cassidy Laidlaw , Jacob Steinhardt

Instruction Fine-Tuning enhances pre-trained language models from basic next-word prediction to complex instruction-following. However, existing One-off Instruction Fine-Tuning (One-off IFT) method, applied on a diverse instruction, may not…

Computation and Language · Computer Science 2024-06-18 Wei Pang , Chuan Zhou , Xiao-Hua Zhou , Xiaojie Wang

Large language models are typically post-trained using supervised fine-tuning (SFT) and reinforcement learning (RL), yet effectively unifying efficient knowledge injection with robust generalization remains challenging. In this work, we…

Artificial Intelligence · Computer Science 2026-05-05 Wangjie Gan , Miao Pan , Linbo Xi , Wenqi Zhang , Jintao Chen , Jianwei Yin , Xuhong Zhang

Supervised fine-tuning (SFT) on chain-of-thought data is an essential post-training step for reasoning language models. Standard machine learning intuition suggests that training with more unique training samples yields better…

Computation and Language · Computer Science 2026-02-12 Dawid J. Kopiczko , Sagar Vaze , Tijmen Blankevoort , Yuki M. Asano

Modern foundation models often undergo iterative ``bootstrapping'' in their post-training phase: a model generates synthetic data, an external verifier filters out low-quality samples, and the high-quality subset is used for further…

Machine Learning · Computer Science 2025-10-17 Pu Yang , Yunzhen Feng , Ziyuan Chen , Yuhang Wu , Zhuoyuan Li

This study evaluates Direct Preference Optimization (DPO) and its variants for aligning Large Language Models (LLMs) with human preferences, testing three configurations: (1) with Supervised Fine Tuning (SFT), (2) without SFT, and (3)…

Computation and Language · Computer Science 2025-02-11 Amir Saeidi , Shivanshu Verma , Md Nayem Uddin , Chitta Baral

Supervised fine-tuning (SFT) is a critical step in aligning large language models (LLMs) with human instructions and values, yet many aspects of SFT remain poorly understood. We trained a wide range of base models on a variety of datasets…

Computation and Language · Computer Science 2025-10-31 Yuto Harada , Yusuke Yamauchi , Yusuke Oda , Yohei Oseki , Yusuke Miyao , Yu Takagi

Conventionally, supervised fine-tuning (SFT) is treated as a simple imitation learning process that only trains a policy to imitate expert behavior on demonstration datasets. In this work, we challenge this view by establishing a…

Machine Learning · Computer Science 2025-10-06 Jiangnan Li , Thuy-Trang Vu , Ehsan Abbasnejad , Gholamreza Haffari

It is common practice in deep learning to use overparameterized networks and train for as long as possible; there are numerous studies that show, both theoretically and empirically, that such practices surprisingly do not unduly harm the…

Machine Learning · Computer Science 2020-03-05 Leslie Rice , Eric Wong , J. Zico Kolter

In inductive transfer learning, fine-tuning pre-trained convolutional networks substantially outperforms training from scratch. When using fine-tuning, the underlying assumption is that the pre-trained model extracts generic features, which…

Machine Learning · Computer Science 2018-06-07 Xuhong Li , Yves Grandvalet , Franck Davoine

Aligning human preference and value is an important requirement for contemporary foundation models. State-of-the-art techniques such as Reinforcement Learning from Human Feedback (RLHF) often consist of two stages: 1) supervised fine-tuning…

Artificial Intelligence · Computer Science 2024-10-29 Jiaxiang Li , Siliang Zeng , Hoi-To Wai , Chenliang Li , Alfredo Garcia , Mingyi Hong