One Model for All Tasks: Leveraging Efficient World Models in Multi-Task Planning

• URL: http://arxiv.org/abs/2509.07945v2
• Date: Thu, 25 Sep 2025 11:04:15 GMT
• Title: One Model for All Tasks: Leveraging Efficient World Models in Multi-Task Planning
• Authors: Yuan Pu, Yazhe Niu, Jia Tang, Junyu Xiong, Shuai Hu, Hongsheng Li,
• Abstract summary: Multi-task world models like UniZero excel in single-task settings.<n>We find that gradient conflicts and the loss of model plasticity often constrain their sample efficiency.<n>In this work, we address these challenges from two complementary perspectives: the single learning iteration and the overall learning process.
• Score: 32.13266149565313
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In heterogeneous multi-task decision-making, tasks not only exhibit diverse observation and action spaces but also vary substantially in their underlying complexities. While conventional multi-task world models like UniZero excel in single-task settings, we find that when handling a broad and diverse suite of tasks, gradient conflicts and the loss of model plasticity often constrain their sample efficiency. In this work, we address these challenges from two complementary perspectives: the single learning iteration and the overall learning process. First, to mitigate the gradient conflicts, we systematically investigate key architectural designs for extending UniZero. Our investigation identifies a Mixture-of-Experts (MoE) architecture as the most effective approach. We demonstrate, both theoretically and empirically, that this architecture alleviates gradient conflicts by routing task-specific representations to specialized sub-networks. This finding leads to our proposed model, \textit{ScaleZero}. Second, to dynamically allocate model capacity throughout the learning process, we introduce an online Dynamic Parameter Scaling (DPS) strategy. This strategy progressively integrates LoRA adapters in response to task-specific progress, enabling adaptive knowledge retention and parameter expansion. Evaluations on a diverse set of standard benchmarks (Atari, DMC, Jericho) demonstrate that ScaleZero, utilizing solely online reinforcement learning with one model, performs on par with specialized single-task agents. With the DPS strategy, it remains competitive while using just 71.5% of the environment interactions. These findings underscore the potential of ScaleZero for effective multi-task planning. Our code is available at \textcolor{magenta}{https://github.com/opendilab/LightZero}.

Related papers

• Model Merging in the Essential Subspace [78.5390284258307]
  Model merging aims to integrate multiple task-specific fine-tuned models into a single multi-task model without additional training.<n>Despite extensive research, task interference remains a major obstacle that often undermines the performance of merged models.<n>We propose ESM (Essential Subspace Merging), a robust framework for effective model merging.
  arXiv  Detail & Related papers  (2026-02-23T00:33:38Z)
• LLaVA-CMoE: Towards Continual Mixture of Experts for Large Vision-Language Models [21.888139819188105]
  LLaVA-CMoE is a continual learning framework for large language models.<n> Probe-Guided Knowledge Extension mechanism determines when and where new experts should be added.<n>Probabilistic Task Locator assigns each task a dedicated, lightweight router.
  arXiv  Detail & Related papers  (2025-03-27T07:36:11Z)
• Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent [72.10987117380584]
  Merging multiple expert models offers a promising approach for performing multi-task learning without accessing their original data.<n>We find existing methods discard task-specific information that, while causing conflicts, is crucial for performance.<n>Our approach consistently outperforms previous methods, achieving state-of-the-art results across diverse architectures and tasks in both vision and NLP domains.
  arXiv  Detail & Related papers  (2025-01-02T12:45:21Z)
• On Giant's Shoulders: Effortless Weak to Strong by Dynamic Logits Fusion [23.63688816017186]
  Existing weak-to-strong methods often employ a static knowledge transfer ratio and a single small model for transferring complex knowledge. We propose a dynamic logit fusion approach that works with a series of task-specific small models, each specialized in a different task. Our method closes the performance gap by 96.4% in single-task scenarios and by 86.3% in multi-task scenarios.
  arXiv  Detail & Related papers  (2024-06-17T03:07:41Z)
• InterroGate: Learning to Share, Specialize, and Prune Representations for Multi-task Learning [17.66308231838553]
  We propose a novel multi-task learning (MTL) architecture designed to mitigate task interference while optimizing inference computational efficiency. We employ a learnable gating mechanism to automatically balance the shared and task-specific representations while preserving the performance of all tasks.
  arXiv  Detail & Related papers  (2024-02-26T18:59:52Z)
• Concrete Subspace Learning based Interference Elimination for Multi-task Model Fusion [86.6191592951269]
  Merging models fine-tuned from common extensively pretrained large model but specialized for different tasks has been demonstrated as a cheap and scalable strategy to construct a multitask model that performs well across diverse tasks. We propose the CONtinuous relaxation dis (Concrete) subspace learning method to identify a common lowdimensional subspace and utilize its shared information track interference problem without sacrificing performance.
  arXiv  Detail & Related papers  (2023-12-11T07:24:54Z)
• An Efficient General-Purpose Modular Vision Model via Multi-Task Heterogeneous Training [79.78201886156513]
  We present a model that can perform multiple vision tasks and can be adapted to other downstream tasks efficiently. Our approach achieves comparable results to single-task state-of-the-art models and demonstrates strong generalization on downstream tasks.
  arXiv  Detail & Related papers  (2023-06-29T17:59:57Z)
• M$^3$ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design [95.41238363769892]
  Multi-task learning (MTL) encapsulates multiple learned tasks in a single model and often lets those tasks learn better jointly. Current MTL regimes have to activate nearly the entire model even to just execute a single task. We present a model-accelerator co-design framework to enable efficient on-device MTL.
  arXiv  Detail & Related papers  (2022-10-26T15:40:24Z)
• Task Adaptive Parameter Sharing for Multi-Task Learning [114.80350786535952]
  Adaptive Task Adapting Sharing (TAPS) is a method for tuning a base model to a new task by adaptively modifying a small, task-specific subset of layers. Compared to other methods, TAPS retains high accuracy on downstream tasks while introducing few task-specific parameters. We evaluate our method on a suite of fine-tuning tasks and architectures (ResNet, DenseNet, ViT) and show that it achieves state-of-the-art performance while being simple to implement.
  arXiv  Detail & Related papers  (2022-03-30T23:16:07Z)
• Controllable Dynamic Multi-Task Architectures [92.74372912009127]
  We propose a controllable multi-task network that dynamically adjusts its architecture and weights to match the desired task preference as well as the resource constraints. We propose a disentangled training of two hypernetworks, by exploiting task affinity and a novel branching regularized loss, to take input preferences and accordingly predict tree-structured models with adapted weights.
  arXiv  Detail & Related papers  (2022-03-28T17:56:40Z)
• Rethinking Hard-Parameter Sharing in Multi-Task Learning [20.792654758645302]
  Hard parameter sharing in multi-task learning (MTL) allows tasks to share some of model parameters, reducing storage cost and improving prediction accuracy. The common sharing practice is to share bottom layers of a deep neural network among tasks while using separate top layers for each task. Using separate bottom-layer parameters could achieve significantly better performance than the common practice.
  arXiv  Detail & Related papers  (2021-07-23T17:26:40Z)
• HyperGrid: Efficient Multi-Task Transformers with Grid-wise Decomposable Hyper Projections [96.64246471034195]
  We propose textscHyperGrid, a new approach for highly effective multi-task learning. Our method helps bridge the gap between fine-tuning and multi-task learning approaches.
  arXiv  Detail & Related papers  (2020-07-12T02:49:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.