Cover Image

SOUP: Token-level Single-sample Mix-policy Reinforcement Learning for Large Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yang, Lei, Bi, Wei, Sun, Chenxi, Jin, Renren, Xiong, Deyi
Format:	Preprint
Published:	2026
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Do Large Language Models Mirror Cognitive Language Processing?
by: Ren, Yuqi, et al.
Published: (2024)

LHMKE: A Large-scale Holistic Multi-subject Knowledge Evaluation Benchmark for Chinese Large Language Models
by: Liu, Chuang, et al.
Published: (2024)

ProBench: Benchmarking Large Language Models in Competitive Programming
by: Yang, Lei, et al.
Published: (2025)

Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models
by: Shi, Dan, et al.
Published: (2026)

FineMath: A Fine-Grained Mathematical Evaluation Benchmark for Chinese Large Language Models
by: Liu, Yan, et al.
Published: (2024)

A Comprehensive Evaluation of Quantization Strategies for Large Language Models
by: Jin, Renren, et al.
Published: (2024)

Revisiting Entropy in Reinforcement Learning for Large Reasoning Models
by: Jin, Renren, et al.
Published: (2025)

Multilingual Large Language Models: A Systematic Survey
by: Zhu, Shaolin, et al.
Published: (2024)

FuxiTranyu: A Multilingual Large Language Model Trained with Balanced Data
by: Sun, Haoran, et al.
Published: (2024)

ConTrans: Weak-to-Strong Alignment Engineering via Concept Transplantation
by: Dong, Weilong, et al.
Published: (2024)

Data Mixing for Large Language Models Pretraining: A Survey and Outlook
by: Chen, Zhuo, et al.
Published: (2026)

Mix-MoE: Improving Multilingual Machine Translation of Large Language Models through Mixed MoEs
by: Li, Bo, et al.
Published: (2026)

CRiskEval: A Chinese Multi-Level Risk Evaluation Benchmark Dataset for Large Language Models
by: Shi, Ling, et al.
Published: (2024)

RoleEval: A Bilingual Role Evaluation Benchmark for Large Language Models
by: Shen, Tianhao, et al.
Published: (2023)

IRCAN: Mitigating Knowledge Conflicts in LLM Generation via Identifying and Reweighting Context-Aware Neurons
by: Shi, Dan, et al.
Published: (2024)

Efficiently Exploring Large Language Models for Document-Level Machine Translation with In-context Learning
by: Cui, Menglong, et al.
Published: (2024)

LFED: A Literary Fiction Evaluation Dataset for Large Language Models
by: Yu, Linhao, et al.
Published: (2024)

From Curated Data to Scalable Models: Continual Pre-training of Dense and MoE Large Language Models for Tibetan
by: Yang, Lei, et al.
Published: (2025)

Joint Training And Decoding for Multilingual End-to-End Simultaneous Speech Translation
by: Huang, Wuwei, et al.
Published: (2025)

Large Language Model Safety: A Holistic Survey
by: Shi, Dan, et al.
Published: (2024)

Self-Pluralising Culture Alignment for Large Language Models
by: Xu, Shaoyang, et al.
Published: (2024)

Problematic Tokens: Tokenizer Bias in Large Language Models
by: Yang, Jin, et al.
Published: (2024)

Star-Agents: Automatic Data Optimization with LLM Agents for Instruction Tuning
by: Zhou, Hang, et al.
Published: (2024)

Evaluating Discourse Cohesion in Pre-trained Language Models
by: He, Jie, et al.
Published: (2025)

Token-level Accept or Reject: A Micro Alignment Approach for Large Language Models
by: Zhang, Yang, et al.
Published: (2025)

A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL
by: Yang, Lei, et al.
Published: (2026)

Evaluating and Improving Graph to Text Generation with Large Language Models
by: He, Jie, et al.
Published: (2025)

FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation
by: Zhu, Shaolin, et al.
Published: (2025)

Exploring Multilingual Concepts of Human Value in Large Language Models: Is Value Alignment Consistent, Transferable and Controllable across Languages?
by: Xu, Shaoyang, et al.
Published: (2024)

MoE-CT: A Novel Approach For Large Language Models Training With Resistance To Catastrophic Forgetting
by: Li, Tianhao, et al.
Published: (2024)

Meta-RTL: Reinforcement-Based Meta-Transfer Learning for Low-Resource Commonsense Reasoning
by: Fu, Yu, et al.
Published: (2024)

TaP: A Taxonomy-Guided Framework for Automated and Scalable Preference Data Generation
by: Jin, Renren, et al.
Published: (2025)

Rethinking Tokenization: Crafting Better Tokenizers for Large Language Models
by: Yang, Jinbiao
Published: (2024)

Iter-AHMCL: Alleviate Hallucination for Large Language Model via Iterative Model-level Contrastive Learning
by: Wu, Huiwen, et al.
Published: (2024)

Towards Understanding Multi-Task Learning (Generalization) of LLMs via Detecting and Exploring Task-Specific Neurons
by: Leng, Yongqi, et al.
Published: (2024)

Large Language Model as Token Compressor and Decompressor
by: Li, Wenbing, et al.
Published: (2026)

Preference-grounded Token-level Guidance for Language Model Fine-tuning
by: Yang, Shentao, et al.
Published: (2023)

Tokens for Learning, Tokens for Unlearning: Mitigating Membership Inference Attacks in Large Language Models via Dual-Purpose Training
by: Tran, Toan, et al.
Published: (2025)

Explore Data Left Behind in Reinforcement Learning for Reasoning Language Models
by: Liu, Chenxi, et al.
Published: (2025)

Reinforcement Learning with Token-level Feedback for Controllable Text Generation
by: Li, Wendi, et al.
Published: (2024)