Cover Image

Demystifying Design Choices of Reinforcement Fine-tuning: A Batched Contextual Bandit Learning Perspective

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Xie, Hong, Hu, Xiao, Tan, Tao, Gu, Haoran, Li, Xin, Han, Jianyu, Lian, Defu, Chen, Enhong
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Rethinking Reinforcement fine-tuning of LLMs: A Multi-armed Bandit Learning Perspective
by: Hu, Xiao, et al.
Published: (2026)

Multiple-play Stochastic Bandits with Prioritized Arm Capacity Sharing
by: Xie, Hong, et al.
Published: (2025)

LLM Cache Bandit Revisited: Addressing Query Heterogeneity for Cost-Effective LLM Inference
by: Yang, Hantao, et al.
Published: (2025)

Federated Contextual Cascading Bandits with Asynchronous Communication and Heterogeneous Users
by: Yang, Hantao, et al.
Published: (2024)

Scaling Federated Linear Contextual Bandits via Sketching
by: Yang, Hantao, et al.
Published: (2026)

Multi-agent Multi-armed Bandits with Stochastic Sharable Arm Capacities
by: Xie, Hong, et al.
Published: (2024)

Model Specific Task Similarity for Vision Language Model Selection via Layer Conductance
by: Yang, Wei, et al.
Published: (2026)

Batched Nonparametric Contextual Bandits
by: Jiang, Rong, et al.
Published: (2024)

AgentCAT: An LLM Agent for Extracting and Analyzing Catalytic Reaction Data from Chemical Engineering Literature
by: Yang, Wei, et al.
Published: (2026)

Analytical and Empirical Study of Herding Effects in Recommendation Systems
by: Xie, Hong, et al.
Published: (2024)

Is Softmax Loss All You Need? A Principled Analysis of Softmax-family Loss
by: Pu, Yuanhao, et al.
Published: (2026)

Beyond Surrogates: A Quantitative Analysis for Inter-Metric Relationships
by: Pu, Yuanhao, et al.
Published: (2026)

Refine Large Language Model Fine-tuning via Instruction Vector
by: Jiang, Gangwei, et al.
Published: (2024)

IBCB: Efficient Inverse Batched Contextual Bandit for Behavioral Evolution History
by: Xu, Yi, et al.
Published: (2024)

Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving
by: Lian, Zhexi, et al.
Published: (2026)

Demystifying Instruction Mixing for Fine-tuning Large Language Models
by: Wang, Renxi, et al.
Published: (2023)

Batch Augmentation with Unimodal Fine-tuning for Multimodal Learning
by: Kabir, H M Dipu, et al.
Published: (2025)

Efficient Machine Unlearning via Influence Approximation
by: Liu, Jiawei, et al.
Published: (2025)

Securing Recommender System via Cooperative Training
by: Wang, Qingyang, et al.
Published: (2024)

Batched Online Contextual Sparse Bandits with Sequential Inclusion of Features
by: Swiers, Rowan, et al.
Published: (2024)

Fine-tuning with Very Large Dropout
by: Zhang, Jianyu, et al.
Published: (2024)

Contextual Rollout Bandits for Reinforcement Learning with Verifiable Rewards
by: Lu, Xiaodong, et al.
Published: (2026)

Breaking Determinism: Fuzzy Modeling of Sequential Recommendation Using Discrete State Space Diffusion Model
by: Xie, Wenjia, et al.
Published: (2024)

Learning to Substitute Components for Compositional Generalization
by: Li, Zhaoyi, et al.
Published: (2025)

Multi-Level Contextual Token Relation Modeling for Machine-Generated Text Detection
by: Wu, Chenwang, et al.
Published: (2026)

Learning Deep Tree-based Retriever for Efficient Recommendation: Theory and Method
by: Liu, Ze, et al.
Published: (2024)

Contextual Bandit with Herding Effects: Algorithms and Recommendation Applications
by: Xu, Luyue, et al.
Published: (2024)

MDAP: A Multi-view Disentangled and Adaptive Preference Learning Framework for Cross-Domain Recommendation
by: Tong, Junxiong, et al.
Published: (2024)

Graph Contextual Reinforcement Learning for Efficient Directed Controller Synthesis
by: Ubukata, Toshihide, et al.
Published: (2025)

Learning from Emptiness: De-biasing Listwise Rerankers with Content-Agnostic Probability Calibration
by: Lv, Hang, et al.
Published: (2026)

Designing an Interpretable Interface for Contextual Bandits
by: Maher, Andrew, et al.
Published: (2024)

Scaling Reasoning Hop Exposes Weaknesses: Demystifying and Improving Hop Generalization in Large Language Models
by: Li, Zhaoyi, et al.
Published: (2026)

CeProAgents: A Hierarchical Agents System for Automated Chemical Process Development
by: Yang, Yuhang, et al.
Published: (2026)

Robust Batched Bandits
by: Guo, Yunwen, et al.
Published: (2025)

Contextual Restless Multi-Armed Bandits with Application to Demand Response Decision-Making
by: Chen, Xin, et al.
Published: (2024)

Foundations and Frontiers of Graph Learning Theory
by: Huang, Yu, et al.
Published: (2024)

Variance-Aware Regret Bounds for Stochastic Contextual Dueling Bandits
by: Di, Qiwei, et al.
Published: (2023)

On the Non-decoupling of Supervised Fine-tuning and Reinforcement Learning in Post-training
by: Niu, Xueyan, et al.
Published: (2026)

A Unified Framework for Adaptive Representation Enhancement and Inversed Learning in Cross-Domain Recommendation
by: Zhang, Luankang, et al.
Published: (2024)

Semantic-preserved Augmentation with Confidence-weighted Fine-tuning for Aspect Category Sentiment Analysis
by: Chai, Yaping, et al.
Published: (2025)