Imagen de Portada

MM-Doc-R1: Training Agents for Long Document Visual Question Answering through Multi-turn Reinforcement Learning

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Lin, Jiahang, Hu, Kai, Wang, Binghai, Zhou, Yuhao, Xi, Zhiheng, Guo, Honglin, Liu, Shichun, Wang, Junzhe, Dou, Shihan, Zhou, Enyu, Yan, Hang, Han, Zhenhua, Gui, Tao, Zhang, Qi, Huang, Xuanjing
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
por: Lin, Jiahang, et al.
Publicado: (2026)

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

Steering LLMs via Scalable Interactive Oversight
por: Zhou, Enyu, et al.
Publicado: (2026)

Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learning
por: Xi, Zhiheng, et al.
Publicado: (2024)

RMB: Comprehensively Benchmarking Reward Models in LLM Alignment
por: Zhou, Enyu, et al.
Publicado: (2024)

Enhancing LLM-based Search Agents via Contribution Weighted Group Relative Policy Optimization
por: Wang, Junzhe, et al.
Publicado: (2026)

StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback
por: Dou, Shihan, et al.
Publicado: (2024)

DocTabQA: Answering Questions from Long Documents Using Tables
por: Wang, Haochen, et al.
Publicado: (2024)

Subspace Defense: Discarding Adversarial Perturbations by Learning a Subspace for Clean Signals
por: Zheng, Rui, et al.
Publicado: (2024)

JFTA-Bench: Evaluate LLM's Ability of Tracking and Analyzing Malfunctions Using Fault Trees
por: Wang, Yuhui, et al.
Publicado: (2026)

LoRAMoE: Alleviate World Knowledge Forgetting in Large Language Models via MoE-Style Plugin
por: Dou, Shihan, et al.
Publicado: (2023)

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

OctoBench: Benchmarking Scaffold-Aware Instruction Following in Repository-Grounded Agentic Coding
por: Ding, Deming, et al.
Publicado: (2026)

DocVXQA: Context-Aware Visual Explanations for Document Question Answering
por: Souibgui, Mohamed Ali, et al.
Publicado: (2025)

Why Reinforcement Fine-Tuning Enables MLLMs Preserve Prior Knowledge Better: A Data Perspective
por: Zhang, Zhihao, et al.
Publicado: (2025)

AdaDocVQA: Adaptive Framework for Long Document Visual Question Answering in Low-Resource Settings
por: Li, Haoxuan, et al.
Publicado: (2025)

DocVideoQA: Towards Comprehensive Understanding of Document-Centric Videos through Question Answering
por: Wang, Haochen, et al.
Publicado: (2025)

DocFusion: A Unified Framework for Document Parsing Tasks
por: Chai, Mingxu, et al.
Publicado: (2024)

BoundingDocs: a Unified Dataset for Document Question Answering with Spatial Annotations
por: Giovannini, Simone, et al.
Publicado: (2025)

Self-Polish: Enhance Reasoning in Large Language Models via Problem Refinement
por: Xi, Zhiheng, et al.
Publicado: (2023)

Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control
por: Zhang, Jiazheng, et al.
Publicado: (2026)

DocPrune:Efficient Document Question Answering via Background, Question, and Comprehension-aware Token Pruning
por: Choi, Joonmyung, et al.
Publicado: (2026)

DocSage: An Information Structuring Agent for Multi-Doc Multi-Entity Question Answering
por: Lin, Teng, et al.
Publicado: (2026)

Self-Demos: Eliciting Out-of-Demonstration Generalizability in Large Language Models
por: He, Wei, et al.
Publicado: (2024)

Secrets of RLHF in Large Language Models Part II: Reward Modeling
por: Wang, Binghai, et al.
Publicado: (2024)

MM-PhyRLHF: Reinforcement Learning Framework for Multimodal Physics Question-Answering
por: Kapuriya, Janak, et al.
Publicado: (2024)

Zero-Shot Complex Question-Answering on Long Scientific Documents
por: Wang, Wanting
Publicado: (2025)

60 Data Points are Sufficient to Fine-Tune LLMs for Question-Answering
por: Ye, Junjie, et al.
Publicado: (2024)

MetaRM: Shifted Distributions Alignment via Meta-Learning
por: Dou, Shihan, et al.
Publicado: (2024)

LLMEval-Med: A Real-world Clinical Benchmark for Medical LLMs with Physician Validation
por: Zhang, Ming, et al.
Publicado: (2025)

VRPO: Rethinking Value Modeling for Robust RL Training under Noisy Supervision
por: Zhu, Dingwei, et al.
Publicado: (2025)

BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping
por: Xi, Zhiheng, et al.
Publicado: (2025)

ASTRA-QA: A Benchmark for Abstract Question Answering over Documents
por: Wang, Shu, et al.
Publicado: (2026)

DocScope: Benchmarking Verifiable Reasoning for Trustworthy Long-Document Understanding
por: Feng, Xiang, et al.
Publicado: (2026)

AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and Progress
por: Xi, Zhiheng, et al.
Publicado: (2025)

SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance
por: Huang, Caishuang, et al.
Publicado: (2024)

MovieChat+: Question-aware Sparse Memory for Long Video Question Answering
por: Song, Enxin, et al.
Publicado: (2024)

LLMEval-Fair: A Large-Scale Longitudinal Study on Robust and Fair Evaluation of Large Language Models
por: Zhang, Ming, et al.
Publicado: (2025)

Improving RL Exploration for LLM Reasoning through Retrospective Replay
por: Dou, Shihan, et al.
Publicado: (2025)

DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training
por: Zhu, Dingwei, et al.
Publicado: (2025)