English
Related papers

Related papers: Positive-Unlabeled Reinforcement Learning Distilla…

200 papers

Current LLM unlearning methods are not robust. A few steps of finetuning can revert their effects. We begin by showing that this is true even for an idealized form of unlearning: training to imitate a model that was never trained on…

In learning-to-rank problems, a privileged feature is one that is available during model training, but not available at test time. Such features naturally arise in merchandised recommendation systems; for instance, "user clicked this item"…

Machine Learning · Computer Science 2022-09-20 Shuo Yang , Sujay Sanghavi , Holakou Rahmanian , Jan Bakus , S. V. N. Vishwanathan

Training-time privileged information (PI) can enable language models to succeed on tasks they would otherwise fail, making it a powerful tool for reinforcement learning in hard, long-horizon settings. However, transferring capabilities…

Machine Learning · Computer Science 2026-02-17 Emiliano Penaloza , Dheeraj Vattikonda , Nicolas Gontier , Alexandre Lacoste , Laurent Charlin , Massimo Caccia

We propose Algorithm Distillation (AD), a method for distilling reinforcement learning (RL) algorithms into neural networks by modeling their training histories with a causal sequence model. Algorithm Distillation treats learning to…

Large language models trained with reinforcement learning (RL) for mathematical reasoning face a fundamental challenge: on problems the model cannot solve at all - "cliff" prompts - the RL gradient vanishes entirely, preventing any learning…

Machine Learning · Computer Science 2026-03-26 Ken Ding

As Large Language Models (LLMs) continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become a central engineering problem, and knowledge distillation remains the dominant…

Machine Learning · Computer Science 2026-05-19 Mingyang Song , Mao Zheng

While Pairwise Ranking Prompting (PRP) with Large Language Models (LLMs) is one of the most effective zero-shot document ranking methods, it has a quadratic computational complexity with respect to the number of documents to be ranked, as…

Information Retrieval · Computer Science 2025-07-08 Junru Wu , Le Yan , Zhen Qin , Honglei Zhuang , Paul Suganthan G. C. , Tianqi Liu , Zhe Dong , Xuanhui Wang , Harrie Oosterhuis

Many attempts have been done to extend the great success of convolutional neural networks (CNNs) achieved on high-end GPU servers to portable devices such as smart phones. Providing compression and acceleration service of deep learning…

Machine Learning · Computer Science 2019-10-09 Yixing Xu , Yunhe Wang , Hanting Chen , Kai Han , Chunjing Xu , Dacheng Tao , Chang Xu

In modern large language models (LLMs), LLM alignment is of crucial importance and is typically achieved through methods such as reinforcement learning from human feedback (RLHF) and direct preference optimization (DPO). However, in most…

Computation and Language · Computer Science 2025-07-24 Songming Zhang , Xue Zhang , Tong Zhang , Bojie Hu , Yufeng Chen , Jinan Xu

Model distillation has emerged as a prominent technique to improve neural search models. To date, distillation taken an offline approach, wherein a new neural model is trained to predict relevance scores between arbitrary queries and…

Information Retrieval · Computer Science 2023-06-19 Sean MacAvaney , Xi Wang

On-policy distillation offers dense, per-token supervision for training reasoning models; however, it remains unclear under which conditions this signal is beneficial and under which it is detrimental. Which teacher model should be used,…

Offline reinforcement learning (RL) methods harness previous experiences to derive an optimal policy, forming the foundation for pre-trained large-scale models (PLMs). When encountering tasks not seen before, PLMs often utilize several…

Machine Learning · Computer Science 2024-11-05 Shengchao Hu , Wanru Zhao , Weixiong Lin , Li Shen , Ya Zhang , Dacheng Tao

On-policy distillation (OPD), which samples trajectories from the student model and supervises them with a teacher at the token level, avoids relying solely on verifiable terminal rewards and can yield better generalization than off-policy…

In reinforcement learning, domain randomisation is an increasingly popular technique for learning more general policies that are robust to domain-shifts at deployment. However, naively aggregating information from randomised domains may…

Machine Learning · Computer Science 2020-12-10 Chenyang Zhao , Timothy Hospedales

Direct Preference Optimization (DPO) is a powerful paradigm to align language models with human preferences using pairwise comparisons. However, its binary win-or-loss supervision often proves insufficient for training small models with…

Computation and Language · Computer Science 2025-09-23 Minchan Kwon , Junwon Ko , Kangil Kim , Junmo Kim

Vision-based deep reinforcement learning (RL) typically obtains performance benefit by using high capacity and relatively large convolutional neural networks (CNN). However, a large network leads to higher inference costs (power, latency,…

Machine Learning · Computer Science 2019-05-01 Sam Green , Craig M. Vineyard , Çetin Kaya Koç

On-policy distillation is pivotal for transferring reasoning capabilities to capacity-constrained models, yet remains prone to instability and negative transfer. We show that on-policy distillation can be interpreted, both theoretically and…

Machine Learning · Computer Science 2026-03-13 Jongwoo Ko , Sara Abdali , Young Jin Kim , Tianyi Chen , Pashmina Cameron

Reinforcement learning (RL) has emerged as a central paradigm for post-training LLM agents, yet its trajectory-level reward signal provides only coarse supervision for long-horizon interaction. On-Policy Self-Distillation (OPSD) complements…

Machine Learning · Computer Science 2026-05-15 Zhengxi Lu , Zhiyuan Yao , Zhuowen Han , Zi-Han Wang , Jinyang Wu , Qi Gu , Xunliang Cai , Weiming Lu , Jun Xiao , Yueting Zhuang , Yongliang Shen

Reinforcement Learning (RL) algorithms for safety alignment of Large Language Models (LLMs), such as Direct Preference Optimization (DPO), encounter the challenge of distribution shift. Current approaches typically address this issue…

Computation and Language · Computer Science 2025-06-17 Qiyuan Deng , Xuefeng Bai , Kehai Chen , Yaowei Wang , Liqiang Nie , Min Zhang

We address the problem of fine-tuning diffusion models for reward-guided generation in biomolecular design. While diffusion models have proven highly effective in modeling complex, high-dimensional data distributions, real-world…