Cover Image

Omni-Thinker: Scaling Multi-Task RL in LLMs with Hybrid Reward and Task Scheduling

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Derek, Zhou, Jiaming, Brunswic, Leo Maxime, Ghaddar, Abbas, Sun, Qianyi, Ma, Liheng, Luo, Yu, Li, Dong, Coates, Mark, Hao, Jianye, Zhang, Yingxue
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Enhancing Logical Reasoning in Large Language Models through Graph-based Synthetic Data
by: Zhou, Jiaming, et al.
Published: (2024)

A Theory of Multi-Agent Generative Flow Networks
by: Brunswic, Leo Maxime, et al.
Published: (2025)

CKGConv: General Graph Convolution with Continuous Kernels
by: Ma, Liheng, et al.
Published: (2024)

Alexandrov Theorem for 2+1 flat radiant spacetimes
by: Brunswic, Léo
Published: (2020)

On branched coverings of singular $(G,X)$-manifolds
by: Brunswic, Léo
Published: (2020)

One Demo Is All It Takes: Planning Domain Derivation with LLMs from A Single Demonstration
by: Huang, Jinbang, et al.
Published: (2025)

Multi-resolution Time-Series Transformer for Long-term Forecasting
by: Zhang, Yitian, et al.
Published: (2023)

GraphPPD: Posterior Predictive Modelling for Graph-Level Inference
by: Pal, Soumyasundar, et al.
Published: (2025)

Plain Transformers Can be Powerful Graph Learners
by: Ma, Liheng, et al.
Published: (2025)

A Theory of Non-Acyclic Generative Flow Networks
by: Brunswic, Leo Maxime, et al.
Published: (2023)

Sparse Decomposition of Graph Neural Networks
by: Hu, Yaochen, et al.
Published: (2024)

OmniEVA: Embodied Versatile Planner via Task-Adaptive 3D-Grounded and Embodiment-aware Reasoning
by: Liu, Yuecheng, et al.
Published: (2025)

On the importance of Data Scale in Pretraining Arabic Language Models
by: Ghaddar, Abbas, et al.
Published: (2024)

BOSCH: Black-Box Binary Optimization for Short-Context Attention-Head Selection in LLMs
by: Ghaddar, Abbas, et al.
Published: (2026)

Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
by: Alomrani, Mohammad Ali, et al.
Published: (2025)

PosterOmni: Generalized Artistic Poster Creation via Task Distillation and Unified Reward Feedback
by: Chen, Sixiang, et al.
Published: (2026)

Beyond Distribution Sharpening: The Importance of Task Rewards
by: Mittal, Sarthak, et al.
Published: (2026)

JPmHC Dynamical Isometry via Orthogonal Hyper-Connections
by: Sengupta, Biswa, et al.
Published: (2026)

Extracting and Following Paths for Robust Relational Reasoning with Large Language Models
by: Zhang, Ge, et al.
Published: (2024)

Generative Models in Decision Making: A Survey
by: Shao, Xinyu, et al.
Published: (2025)

Hybrid Reward Normalization for Process-supervised Non-verifiable Agentic Tasks
by: Xu, Peiran, et al.
Published: (2025)

FEval-TTC: Fair Evaluation Protocol for Test-Time Compute
by: Rumiantsev, Pavel, et al.
Published: (2025)

Refining Answer Distributions for Improved Large Language Model Reasoning
by: Pal, Soumyasundar, et al.
Published: (2024)

A Balanced Neuro-Symbolic Approach for Commonsense Abductive Logic
by: Cotnareanu, Joseph, et al.
Published: (2026)

Personalized Negative Reservoir for Incremental Learning in Recommender Systems
by: Valkanas, Antonios, et al.
Published: (2024)

Learning Reward Machines in Cooperative Multi-Agent Tasks
by: Ardon, Leo, et al.
Published: (2023)

H-WM: Robotic Task and Motion Planning Guided by Hierarchical World Model
by: Huang, Jinbang, et al.
Published: (2026)

Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for Reasoning
by: Mahrooghi, Ilia, et al.
Published: (2026)

uTeBC-NLP at SemEval-2024 Task 9: Can LLMs be Lateral Thinkers?
by: Sadeghi, Pouya, et al.
Published: (2024)

Ergodic Generative Flows
by: Brunswic, Leo Maxime, et al.
Published: (2025)

SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling
by: Zhang, Yiqi, et al.
Published: (2026)

Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning
by: Wu, Yihong, et al.
Published: (2025)

DyG2Vec: Efficient Representation Learning for Dynamic Graphs
by: Alomrani, Mohammad Ali, et al.
Published: (2022)

Exploring RL-based LLM Training for Formal Language Tasks with Programmed Rewards
by: Padula, Alexander G., et al.
Published: (2024)

Parametric Feature Transfer: One-shot Federated Learning with Foundation Models
by: Beitollahi, Mahdi, et al.
Published: (2024)

SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards
by: Batra, Hunar, et al.
Published: (2025)

MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech
by: Wang, Chengyao, et al.
Published: (2025)

Imbalanced Gradients in RL Post-Training of Multi-Task LLMs
by: Wu, Runzhe, et al.
Published: (2025)

When are LLMs Sufficient Policy Optimizers for Sequential RL Tasks?
by: Hatgis-Kessell, Stephane, et al.
Published: (2026)

OmniOVCD: Streamlining Open-Vocabulary Change Detection with SAM 3
by: Zhang, Xu, et al.
Published: (2026)