arXiv Machine Learning

JumpStart Your Policy Learning with Lessons from 160,000 Training Runs

arXiv Machine Learning
Oct 12

Cross-Behavior Policy Optimization: Unified Learning from Offline Demonstrations and On- and Off-Policy Rollouts via a Shared Regression Target

The paper introduces Cross-Behavior Policy Optimization (CBPO), a method that unifies offline demonstrations and policy-generated rollouts into a single regression loss using a fixed-reference target. CBPO eliminates the need for separate supervised fine‑tuning and reinforcement learning objectives, allowing positive and negative offline completions, reference responses, and current and historical rollouts to share a common reward‑labeled format and replay buffer. Experiments on two model scales and four mathematical reasoning benchmarks show that CBPO, especially the CBPO‑Unified variant, outperforms baseline methods in avg@32 and pass@32 metrics, with historical rollouts providing additional gains.

By Du Cheng, Junyu Lu, Jingfan Zhang, Yuzhuo Zhang, Haoxuan Jia, Shichao Weng
arXiv Machine Learning
Sep 14

MInTRL: Off-policy Intervention can boost On-policy RL

MInTRL (Minimal Intervention Reinforcement Learning) expands exploration in on-policy reinforcement learning by inserting sparse, local corrections into rollouts via a judge-intervention policy. These interventions replace erroneous suffixes and immediately return control to the main policy, allowing the agent to explore beyond its natural trajectory while maintaining on-policy data. The method uses a sequence-level advantage-regression objective, avoiding importance sampling, and demonstrates superior performance on math and code benchmarks compared to standard on-policy and off-policy baselines.

By Mingyu Chen, Yefan Tao, Gerald Friedland, Xuezhou Zhang, Chris Kong
arXiv Machine Learning
Sep 18

Improving Online Reinforcement Learning via Bidirectional Behavior Prior Distillation

The paper introduces Bidirectional Behavior Prior Distillation (B2PD), a method that uses action‑value priors to train a conditional variational autoencoder for generating high‑value behavior support. These expert behavior priors are then distilled into the online reinforcement learning agent, reducing inefficient exploration and stabilizing policy updates. Experiments on state‑ and pixel‑based tasks show that B2PD improves sample efficiency while maintaining stable learning dynamics.

By Gong Gao, Xiao Lai, Jiaji Shen, Ning Jia, Xianhui Liu, Weidong Zhao
arXiv Machine Learning
Oct 3

Reusing Past Samples in Proximal Policy Optimization: When and How Does It Help?

The paper investigates how reusing past samples can improve the sample efficiency of Proximal Policy Optimization (PPO). Two variants, wPPO-U and wPPO-BH, are introduced within a multiple importance weighting framework, each reusing data from recent iterations while preserving core PPO mechanics. The authors derive theoretical policy improvement bounds for both variants and empirically evaluate their impact on continuous control tasks.

By Alessandro Montenegro, Riccardo Venturelli, Marco Mussi, Matteo Papini, Alberto Maria Metelli
arXiv AI
Jun 2

Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation

arXiv:2602. 07298v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) represent a promising frontier for recommender systems, yet their development has been impeded by the absence of predictable scaling laws, which are crucial for guiding research and optimizing resource allocation.

By Benyu Zhang, Qiang Zhang, Jianpeng Cheng, Hong-You Chen, Qifei Wang, Wei Sun, Shen Li, Jia Li, Jiahao Wu, Qunshu Zhang, Neeraj Bhatia, Xiangjun Fan, Hong Yan
arXiv Computation and Language
Nov 9

Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters

LambdaMix is a parameter‑efficient method that learns a family of policies conditioned on user preferences by jointly training a low‑rank adapter for each objective over a frozen base model. A single preference vector selects an adapter mixture to generate responses and weight normalized rewards, enabling on‑line reinforcement learning without further optimization at inference. Experiments on four models show LambdaMix outperforms baselines in helpfulness, harmlessness, math problem solving, and comment rewriting, while using fewer training completions and storing fewer adapter parameters.

By Constantinos Karouzos, Nikolaos Aletras