Cover Image

Learning What Reinforcement Learning Can't: Interleaved Online Fine-Tuning for Hardest Questions

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ma, Lu, Liang, Hao, Qiang, Meiyi, Tang, Lexiang, Ma, Xiaochen, Wong, Zhen Hao, Niu, Junbo, Shen, Chengyu, He, Runming, Li, Yanhao, Cui, Bin, Zhang, Wentao
Format:	Preprint
Published:	2025
Subjects:	Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Let's Verify Math Questions Step by Step
by: Shen, Chengyu, et al.
Published: (2025)

K12-KGraph: A Curriculum-Aligned Knowledge Graph for Benchmarking and Training Educational LLMs
by: Liang, Hao, et al.
Published: (2026)

One-Eval: An Agentic System for Automated and Traceable LLM Evaluation
by: Shen, Chengyu, et al.
Published: (2026)

FlipVQA: Scaling Multi-modal Instruction Tuning via Textbook-to-Knowledge Synthesis
by: Wong, Zhen Hao, et al.
Published: (2025)

LogicPuzzleRL: Cultivating Robust Mathematical Reasoning in LLMs via Reinforcement Learning
by: Wong, Zhen Hao, et al.
Published: (2025)

Enhancing Reinforcement Learning Fine-Tuning with an Online Refiner
by: Ma, Hao, et al.
Published: (2026)

Leash: Adaptive Length Penalty and Reward Shaping for Efficient Large Reasoning Model
by: Li, Yanhao, et al.
Published: (2025)

Reinforcement Learning Outperforms Supervised Fine-Tuning: A Case Study on Audio Question Answering
by: Li, Gang, et al.
Published: (2025)

GIFT: Reconciling Post-Training Objectives via Finite-Temperature Gibbs Initialization
by: Zhao, Zhengyang, et al.
Published: (2026)

Online Learning from Strategic Human Feedback in LLM Fine-Tuning
by: Hao, Shugang, et al.
Published: (2024)

IN-RIL: Interleaved Reinforcement and Imitation Learning for Policy Fine-Tuning
by: Gao, Dechen, et al.
Published: (2025)

Climbing the Ladder of Reasoning: What LLMs Can-and Still Can't-Solve after SFT?
by: Sun, Yiyou, et al.
Published: (2025)

DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models
by: Liang, Hao, et al.
Published: (2026)

Supervised Fine-Tuning as Inverse Reinforcement Learning
by: Sun, Hao
Published: (2024)

FLARE: Full-Modality Long-Video Audiovisual Retrieval Benchmark with User-Simulated Queries
by: You, Qijie, et al.
Published: (2026)

Formal Logic Enabled Personalized Federated Learning Through Property Inference
by: An, Ziyan, et al.
Published: (2024)

Autocrats Can't Always Get What They Want
by: Brown, Nathan J., et al.
Published: (2024)

DARO: Difficulty-Aware Reweighting Policy Optimization
by: Zhou, Jingyu, et al.
Published: (2025)

Towards Next-Generation LLM Training: From the Data-Centric Perspective
by: Liang, Hao, et al.
Published: (2026)

They Can't Always Find What They Want; Kids' Online Behaviors Have Researchers Scratching Their Heads
by: Minkel, Walter
Published: (2004)

Reverse Question Answering: Can an LLM Write a Question so Hard (or Bad) that it Can't Answer?
by: Balepur, Nishant, et al.
Published: (2024)

DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI
by: Liang, Hao, et al.
Published: (2025)

Gradual Learning: Optimizing Fine-Tuning with Partially Mastered Knowledge in Large Language Models
by: Li, Bozhou, et al.
Published: (2024)

Can’t Touch This
Published: (2024)

Why I Can't Create a Learning Center
by: Miller, Rosalind
Published: (1975)

Learning with Preserving for Continual Multitask Learning
by: Wang, Hanchen David, et al.
Published: (2025)

Fine-Tuning Language Models with Reward Learning on Policy
by: Lang, Hao, et al.
Published: (2024)

Hardest Monotone Functions for Evolutionary Algorithms
by: Kaufmann, Marc, et al.
Published: (2023)

EVQAScore: A Fine-grained Metric for Video Question Answering Data Quality Evaluation
by: Liang, Hao, et al.
Published: (2024)

Human Resilience in the AI Era -- What Machines Can't Replace
by: Liu, Shaoshan, et al.
Published: (2025)

AbsenceBench: Language Models Can't Tell What's Missing
by: Fu, Harvey Yiyun, et al.
Published: (2025)

MathClean: A Benchmark for Synthetic Mathematical Data Cleaning
by: Liang, Hao, et al.
Published: (2025)

BRACE: A Benchmark for Robust Audio Caption Quality Evaluation
by: Guo, Tianyu, et al.
Published: (2025)

Integrating Offline Pre-Training with Online Fine-Tuning: A Reinforcement Learning Approach for Robot Social Navigation
by: Su, Run, et al.
Published: (2025)

They Can't Hear Us Does Not Mean We Can't Serve Them.
by: McDaniel, Julie Ann
Published: (1992)

QAEncoder: Towards Aligned Representation Learning in Question Answering Systems
by: Wang, Zhengren, et al.
Published: (2024)

Time Blindness: Why Video-Language Models Can't See What Humans Can?
by: Upadhyay, Ujjwal, et al.
Published: (2025)

Humanity's Last Code Exam: Can Advanced LLMs Conquer Human's Hardest Code Competition?
by: Li, Xiangyang, et al.
Published: (2025)

Coevolving with the Other You: Fine-Tuning LLM with Sequential Cooperative Multi-Agent Reinforcement Learning
by: Ma, Hao, et al.
Published: (2024)

Equal Requests are Asymptotically Hardest for Data Recovery
by: Lember, Jüri, et al.
Published: (2024)