Imagen de Portada

Data-adaptive Safety Rules for Training Reward Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Li, Xiaomin, Gao, Mingye, Zhang, Zhiwei, Fan, Jingxuan, Li, Weiyu
Formato:	Preprint
Publicado:	2025
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

ENCORE: Entropy-guided Reward Composition for Multi-head Safety Reward Models
por: Li, Xiaomin, et al.
Publicado: (2025)

Selection of LLM Fine-Tuning Data based on Orthogonal Rules
por: Li, Xiaomin, et al.
Publicado: (2024)

MedGUIDE: Benchmarking Clinical Decision-Making in Large Language Models
por: Li, Xiaomin, et al.
Publicado: (2025)

Training Data Efficiency in Multimodal Process Reward Models
por: Li, Jinyuan, et al.
Publicado: (2026)

Bradley-Terry and Multi-Objective Reward Modeling Are Complementary
por: Zhang, Zhiwei, et al.
Publicado: (2025)

R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning
por: Zhang, Yi-Fan, et al.
Publicado: (2025)

RRM: Robust Reward Model Training Mitigates Reward Hacking
por: Liu, Tianqi, et al.
Publicado: (2024)

SAFER: Probing Safety in Reward Models with Sparse Autoencoder
por: Shi, Wei, et al.
Publicado: (2025)

MetaRuleGPT: Recursive Numerical Reasoning of Language Models Trained with Simple Rules
por: Chen, Kejie, et al.
Publicado: (2024)

OmniCompliance-100K: A Multi-Domain, Rule-Grounded, Real-World Safety Compliance Dataset
por: Hu, Wenbin, et al.
Publicado: (2026)

Training-free Context-adaptive Attention for Efficient Long Context Modeling
por: You, Zeng, et al.
Publicado: (2025)

The Bidirectional Process Reward Model
por: Zhang, Lingyin, et al.
Publicado: (2025)

GRAM-R$^2$: Self-Training Generative Foundation Reward Models for Reward Reasoning
por: Wang, Chenglong, et al.
Publicado: (2025)

A Comprehensive Survey of Accelerated Generation Techniques in Large Language Models
por: Khoshnoodi, Mahsa, et al.
Publicado: (2024)

Retrieval, Reward, and Training Protocols: What Matters in Training Search Agents?
por: Zhao, Yibo, et al.
Publicado: (2026)

Training LLMs for Multi-Step Tool Orchestration with Constrained Data Synthesis and Graduated Rewards
por: Jiayang, Cheng, et al.
Publicado: (2026)

CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learning with Implicit Rule-Based Rewards
por: Liu, Cheng, et al.
Publicado: (2025)

STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules
por: Wu, Di, et al.
Publicado: (2026)

CARES: Comprehensive Evaluation of Safety and Adversarial Robustness in Medical LLMs
por: Chen, Sijia, et al.
Publicado: (2025)

Pre-Trained Policy Discriminators are General Reward Models
por: Dou, Shihan, et al.
Publicado: (2025)

An Efficient and Precise Training Data Construction Framework for Process-supervised Reward Model in Mathematical Reasoning
por: Sun, Wei, et al.
Publicado: (2025)

Wait, but Tylenol is Acetaminophen... Investigating and Improving Language Models' Ability to Resist Requests for Misinformation
por: Chen, Shan, et al.
Publicado: (2024)

Prior Constraints-based Reward Model Training for Aligning Large Language Models
por: Zhou, Hang, et al.
Publicado: (2024)

Limited Preference Data? Learning Better Reward Model with Latent Space Synthesis
por: Tao, Leitian, et al.
Publicado: (2025)

CAST: Continuous and Differentiable Semi-Structured Sparsity-Aware Training for Large Language Models
por: Huang, Weiyu, et al.
Publicado: (2025)

A Survey on Training-free Alignment of Large Language Models
por: Pan, Birong, et al.
Publicado: (2025)

StoryAlign: Evaluating and Training Reward Models for Story Generation
por: Xia, Haotian, et al.
Publicado: (2026)

Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned Model
por: Zhu, Wenhong, et al.
Publicado: (2024)

Tree Reward-Aligned Search for TReASURe in Masked Diffusion Language Models
por: Yu, Zichao, et al.
Publicado: (2025)

Reward-SQL: Boosting Text-to-SQL via Stepwise Reasoning and Process-Supervised Rewards
por: Zhang, Yuxin, et al.
Publicado: (2025)

Aligning Language Models Using Follow-up Likelihood as Reward Signal
por: Zhang, Chen, et al.
Publicado: (2024)

Business as Rulesual: A Benchmark and Framework for Business Rule Flow Modeling with LLMs
por: Yang, Chen, et al.
Publicado: (2025)

MPO: Multilingual Safety Alignment via Reward Gap Optimization
por: Zhao, Weixiang, et al.
Publicado: (2025)

Tool-Augmented Reward Modeling
por: Li, Lei, et al.
Publicado: (2023)

Exploring Scaling Laws for Local SGD in Large Language Model Training
por: He, Qiaozhi, et al.
Publicado: (2024)

Pruning Large Language Models with Semi-Structural Adaptive Sparse Training
por: Huang, Weiyu, et al.
Publicado: (2024)

Prompt-Level Reward Specifications for Open-Ended Post-Training
por: Weng, Zijun, et al.
Publicado: (2026)

Lessons from Training Grounded LLMs with Verifiable Rewards
por: Sim, Shang Hong, et al.
Publicado: (2025)

EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repair
por: Chen, Jiaao, et al.
Publicado: (2026)

Efficient Safety Alignment of Large Language Models via Preference Re-ranking and Representation-based Reward Modeling
por: Deng, Qiyuan, et al.
Publicado: (2025)