English
Related papers

Related papers: SFT-GRPO Data Overlap as a Post-Training Hyperpara…

200 papers

This paper provides a self-contained, from-scratch, exposition of key algorithms for instruction tuning of models: SFT, Rejection Sampling, REINFORCE, Trust Region Policy Optimization (TRPO), Proximal Policy Optimization (PPO), Group…

Computation and Language · Computer Science 2025-10-22 Rohit Patel

Supervised Fine-Tuning (SFT) is the standard approach for adapting large language models (LLMs) to downstream tasks. However, we observe a persistent failure mode: even after convergence, models often fail to correctly reproduce a subset of…

Computation and Language · Computer Science 2026-04-27 Chao Xue , Yao Wang , Mengqiao Liu , Di Liang , Xingsheng Han , Peiyang Liu , Xianjie Wu , Chenyao Lu , Lei Jiang , Yu Lu , Haibo Shi , Shuang Liang , Minlong Peng , Flora D. Salim

Supervised Fine-Tuning (SFT) has been a go-to and effective method for enhancing long chain-of-thought (CoT) reasoning in relatively small LLMs by fine-tuning them with long CoT responses from larger LLMs. To continually improve reasoning…

Machine Learning · Computer Science 2025-02-20 Wang Yang , Hongye Jin , Jingfeng Yang , Vipin Chaudhary , Xiaotian Han

Scaled post-training now drives many of the largest capability gains in language models (LMs), yet its effect on pretrained knowledge remains poorly understood. Not all forgetting is equal: Forgetting one fact (e.g., a U.S. president or an…

Machine Learning · Computer Science 2025-10-21 Jackson Harmon , Andreas Hochlehnert , Matthias Bethge , Ameya Prabhu

Large language models (LLMs) trained via pretraining and supervised fine-tuning (SFT) can still produce harmful and misaligned outputs, or struggle in domains like math and coding. Reinforcement learning (RL)-based post-training methods,…

Computation and Language · Computer Science 2026-05-19 Zhichao Wang , Kiran Ramnath , Bin Bi , Shiva Kumar Pentyala , Sougata Chaudhuri , Shubham Mehrotra , Zixu , Zhu , Xiang-Bo Mao , Sitaram Asur , Na , Cheng

The Superficial Alignment Hypothesis posits that almost all of a language model's abilities and knowledge are learned during pre-training, while post-training is about giving a model the right style and format. We re-examine these claims by…

Computation and Language · Computer Science 2024-10-08 Mohit Raghavendra , Vaskar Nath , Sean Hendryx

Direct Preference Optimization (DPO) is widely used after supervised fine-tuning (SFT) to align language models, yet empirical behavior under small backbones and modest data is under-specified. We systematically compare SFT-only, DPO-only,…

Computation and Language · Computer Science 2026-03-23 Yuming Feng , Christy Yang

Pretrained vision-language models (VLMs) such as CLIP have shown impressive generalization capability in downstream vision tasks with appropriate text prompts. Instead of designing prompts manually, Context Optimization (CoOp) has been…

Computer Vision and Pattern Recognition · Computer Science 2023-02-15 Chengcheng Ma , Yang Liu , Jiankang Deng , Lingxi Xie , Weiming Dong , Changsheng Xu

Recent progress in flow-based generative models and reinforcement learning (RL) has improved text-image alignment and visual quality. However, current RL training for flow models still has two main problems: (i) GRPO-style fixed per-prompt…

Computer Vision and Pattern Recognition · Computer Science 2026-01-13 Kaijie Chen , Zhiyang Xu , Ying Shen , Zihao Lin , Yuguang Yao , Lifu Huang

Group Relative Policy Optimization (GRPO) enhances policy learning by computing gradients from relative comparisons among candidate outputs that share a common input prefix. Despite its effectiveness, GRPO introduces substantial…

Machine Learning · Computer Science 2025-06-09 Zikang Liu , Tongtian Yue , Yepeng Tang , Longteng Guo , Junxian Cai , Qingbin Liu , Xi Chen , Jing Liu

Mathematical reasoning remains a challenging area for large language models (LLMs), prompting the development of math-specific LLMs such as LLEMMA, DeepSeekMath, and Qwen2-Math, among others. These models typically follow a two-stage…

Computation and Language · Computer Science 2025-03-25 Zui Chen , Tianqiao Liu , Mi Tian , Qing Tong , Weiqi Luo , Zitao Liu

As Test-Time Scaling emerges as an active research focus in the large language model community, advanced post-training methods increasingly emphasize extending chain-of-thought (CoT) generation length, thereby enhancing reasoning…

Artificial Intelligence · Computer Science 2025-05-20 Muzhi Dai , Chenxu Yang , Qingyi Si

Post-training has become essential for adapting large language models (LLMs) to complex downstream behaviors, including instruction following, preference alignment, and multi-step reasoning. Reinforcement learning with verifiable rewards…

Machine Learning · Computer Science 2026-05-20 Chengqian Zhang , Wei Zhu , Kyumin Lee

Post-training of reasoning LLMs is a holistic process that typically consists of an offline SFT stage followed by an online reinforcement learning (RL) stage. However, SFT is often optimized in isolation to maximize SFT performance alone.…

Machine Learning · Computer Science 2026-05-29 Dylan Zhang , Yufeng Xu , Haojin Wang , Qingzhi Chen , Hao Peng

A prevailing narrative in LLM post-training holds that supervised finetuning (SFT) memorizes while reinforcement learning (RL) generalizes. We revisit this claim for reasoning SFT with long chain-of-thought (CoT) supervision and find that…

Artificial Intelligence · Computer Science 2026-04-09 Qihan Ren , Peng Wang , Ruikun Cai , Shuai Shao , Dadi Guo , Yuejin Xie , Yafu Li , Quanshi Zhang , Xia Hu , Jing Shao , Dongrui Liu

Multimodal reasoning in vision-language models (VLMs) typically relies on a two-stage process: supervised fine-tuning (SFT) and reinforcement learning (RL). In standard SFT, all tokens contribute equally to the loss, even though reasoning…

Artificial Intelligence · Computer Science 2026-03-20 Shaked Perek , Ben Wiesel , Avihu Dekel , Nimrod Shabtay , Eli Schwartz

Large reasoning models (LRMs) are commonly trained with reinforcement learning (RL) to explore long chain-of-thought reasoning, achieving strong performance at high computational cost. Recent methods add multi-reward objectives to jointly…

Artificial Intelligence · Computer Science 2026-02-17 Anhao Zhao , Ziyang Chen , Junlong Tong , Yingqi Fan , Fanghua Ye , Shuhao Li , Yunpu Ma , Wenjie Li , Xiaoyu Shen

Supervised fine-tuning (SFT) is crucial for adapting Large Language Models (LLMs) to specific tasks. In this work, we demonstrate that the order of training data can lead to significant training imbalances, potentially resulting in…

Computation and Language · Computer Science 2024-10-08 Yiming Ju , Ziyi Ni , Xingrun Xing , Zhixiong Zeng , hanyu Zhao , Siqi Fan , Zheng Zhang

Large language models (LLMs) with enormous pre-training tokens and parameters emerge diverse abilities, including math reasoning, code generation, and instruction following. These abilities are further enhanced by supervised fine-tuning…

Computation and Language · Computer Science 2024-06-10 Guanting Dong , Hongyi Yuan , Keming Lu , Chengpeng Li , Mingfeng Xue , Dayiheng Liu , Wei Wang , Zheng Yuan , Chang Zhou , Jingren Zhou

Group-relative policy optimization methods train language models by generating multiple rollouts per prompt and normalizing rewards with a shared mean reward baseline. In resource-constrained settings where the rollout budget is small,…

Machine Learning · Computer Science 2026-02-02 Youngeun Kim
‹ Prev 1 4 5 6 7 8 10 Next ›