Imagen de Portada

RiskPO: Risk-based Policy Optimization via Verifiable Reward for LLM Post-Training

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Ren, Tao, Jiang, Jinyang, Yang, Hui, Tian, Wan, Zou, Minhao, Li, Guanghao, Zhang, Zishi, Wang, Qinghao, Qin, Shentao, Zhao, Yanjun, Tao, Rui, Shao, Hui, Peng, Yijie
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Nonparametric Bayesian Optimization for General Rewards
por: Zhang, Zishi, et al.
Publicado: (2026)

RiskMiner: Discovering Formulaic Alphas via Risk Seeking Monte Carlo Tree Search
por: Ren, Tao, et al.
Publicado: (2024)

FLOPS: Forward Learning with OPtimal Sampling
por: Ren, Tao, et al.
Publicado: (2024)

Optimal low-rank stochastic gradient estimation for LLM training
por: Li, Zehao, et al.
Publicado: (2026)

Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio Optimizer
por: Ren, Tao, et al.
Publicado: (2025)

Omni-Masked Gradient Descent: Memory-Efficient Optimization via Mask Traversal with Improved Convergence
por: Yang, Hui, et al.
Publicado: (2026)

Sample-Efficient "Clustering and Conquer" Procedures for Parallel Large-Scale Ranking and Selection
por: Zhang, Zishi, et al.
Publicado: (2024)

Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Training
por: Xu, Ran, et al.
Publicado: (2026)

Risk-Controlled Post-Processing of Decision Policies
por: Joshi, Sunay, et al.
Publicado: (2026)

LLM-Inspired Pretrain-Then-Finetune for Small-Data, Large-Scale Optimization
por: Zhang, Zishi, et al.
Publicado: (2026)

Post-Training Local LLM Agents for Linux Privilege Escalation with Verifiable Rewards
por: Normann, Philipp, et al.
Publicado: (2026)

Deep Reinforcement Learning for Solving Management Problems: Towards A Large Management Mode
por: Jiang, Jinyang, et al.
Publicado: (2024)

Stochastic Approximation Methods for Distortion Risk Measure Optimization
por: Jiang, Jinyang, et al.
Publicado: (2025)

GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training
por: Cao, Yuan, et al.
Publicado: (2026)

Quantum Verifiable Rewards for Post-Training Qiskit Code Assistant
por: Dupuis, Nicolas, et al.
Publicado: (2025)

Task-Centric Policy Optimization from Misaligned Motion Priors
por: Zheng, Ziang, et al.
Publicado: (2026)

A Dense Reward View on Aligning Text-to-Image Diffusion with Preference
por: Yang, Shentao, et al.
Publicado: (2024)

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
por: He, Haoran, et al.
Publicado: (2025)

HarmRLVR: Weaponizing Verifiable Rewards for Harmful LLM Alignment
por: Liu, Yuexiao, et al.
Publicado: (2025)

PRISM: A Unified Framework for Post-Training LLMs Without Verifiable Rewards
por: Ghimire, Mukesh, et al.
Publicado: (2026)

Consolidating Rewarded Perturbations for LLM Post-Training
por: Zhang, Zheyu, et al.
Publicado: (2026)

AlphaPO: Reward Shape Matters for LLM Alignment
por: Gupta, Aman, et al.
Publicado: (2025)

What about gravity in video generation? Post-Training Newton's Laws with Verifiable Rewards
por: Le, Minh-Quan, et al.
Publicado: (2025)

EVPO: Explained Variance Policy Optimization for Adaptive Critic Utilization in LLM Post-Training
por: Pan, Chengjun, et al.
Publicado: (2026)

Copper Current Collector: The Cornerstones of Practical Lithium Metal and Anode–Free Batteries
por: Jinyang Zhou, et al.
Publicado: (2024)

Adaptive Robust Estimator for Multi-Agent Reinforcement Learning
por: Li, Zhongyi, et al.
Publicado: (2026)

Convergence of Neural Network Policies for Risk--Reward Optimization
por: Chen, Chang, et al.
Publicado: (2026)

Navigating the Terrain of Post‐Polypectomy Surveillance: Charting the Complex Landscape of Guidelines and Recurrence Risks
por: Xiaojin Tao, et al.
Publicado: (2026)

Hypertensive Uveitis After Intravitreal Faricimab: Clinical and Methodological Considerations
por: Guanghao Qin
Publicado: (2026)

Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards
por: Ren, Mengjie, et al.
Publicado: (2026)

Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training
por: Liu, Yixin, et al.
Publicado: (2026)

Cold-Start Forecasting of New Product Life-Cycles via Conditional Diffusion Models
por: Zhou, Ruihan, et al.
Publicado: (2026)

Reward-Weighted On-Policy Distillation with an Open Property-Equivalence Verifier for NL-to-SVA Generation
por: Zou, Qingyun, et al.
Publicado: (2026)

Keeping an Eye on LLM Unlearning: The Hidden Risk and Remedy
por: Ren, Jie, et al.
Publicado: (2025)

Risk Capacity and Optimal Monetary Policy
por: Sun, Rui
Publicado: (2026)

From Verifiable Dot to Reward Chain: Harnessing Verifiable Reference-based Rewards for Reinforcement Learning of Open-ended Generation
por: Jiang, Yuxin, et al.
Publicado: (2026)

Comment on “Research on the Integrated Training Mode of Higher Art Education for the Deaf”
por: Hui Shao
Publicado: (2023)

Navigating Ecological Risk: Firm‐Level Innovation Responses to Biodiversity Risks in the United States
por: Miaomiao Tao
Publicado: (2026)

A Motif-Based Framework for Decomposing Risk Spillovers
por: Shao, Ying-Hui, et al.
Publicado: (2026)

"Think First, Verify Always": Training Humans to Face AI Risks
por: Aydin, Yuksel
Publicado: (2025)