arXiv subjects

Yiqin Yang

Publications and source records attributed to Yiqin Yang.

At least 19 recordsLinked to original sources

arXiv2026

Graph-Operator World Models for Morphology-Parameter Generalization in Continuous Control

World models for continuous control are commonly trained for a fixed physical system and can degrade when known morphology parameters such as link lengths, masses, damping, and actuation change. Existing approaches often provide these parameters as conditioning information, but leave unspecified which part of the learned transition should remain reusable and which part should change with morphology. We propose Graph-Operator World Models (GraphOp-WM), a structured world model for generalization across unseen morphology parameters within related articulated robot families. GraphOp-WM represents bodies and their kinematic relations as an attributed graph and factorizes each transition into a morphology-independent local dynamics basis and a morphology-conditioned structured operator. The operator combines node-local modulation, kinematic-tree coupling, and a low-rank global correction, while architectural information separation, basis normalization, and paired-morphology supervision encourage static morphology dependence to be carried by the operator pathway. Graph-level readout and edge-wise action representations provide a compatible interface for reward, value, and TD-MPC-style planning. We further define controlled MuJoCo parameter splits covering interpolation, extrapolation, and held-out compositions of link geometry, mass, damping, and actuation parameters in Hopper, Walker2d, and HalfCheetah.

cs.AI↗

arXiv2026

Multi-Agent Reinforcement Learning via Agent-Specific Preference

Multi-agent reinforcement learning (MARL) is a powerful framework for solving complex collaborative tasks, but it relies heavily on well-defined global reward functions. Designing such rewards is challenging, especially in systems with heterogeneous agents, where a single scalar objective may fail to capture diverse behaviors. In this paper, we introduce Multi-AGent Preference-Integrated lEarning (MAGPIE), which addresses these challenges through agent-specific preference modeling. Each agent is evaluated by a dedicated expert through preference signals, eliminating the need for global evaluation. We theoretically prove that optimizing these decentralized preferences converges to a Nash equilibrium policy. To integrate local preferences into a coherent global objective, we construct agent-specific reward models from preference data and combine them via a monotonic aggregation mechanism. We further prove that optimizing this aggregate reward model is equivalent to training the Nash equilibrium policy. Extensive experiments on benchmark multi-agent tasks and a sequential production line task show that MAGPIE achieves performance comparable to reward-engineered baselines, demonstrating its potential to facilitate policy learning in scenarios where precise reward engineering is impractical.

cs.LG↗

arXiv2026

Sim2O: Efficient Offline-to-Online MARL via Joint Action Composition

Offline-to-online adaptation serves as a pivotal paradigm for mitigating the prohibitive cost of online exploration by bootstrapping reinforcement learning from offline datasets. While this paradigm has been extensively studied in single-agent settings, its extension to Multi-Agent Reinforcement Learning (MARL) remains largely unexplored, despite its critical relevance to complex coordinated decision-making. To bridge this gap, we introduce Sim2O, an elegant and minimalist framework for offline-to-online MARL. Rather than treating adaptation as a monolithic joint decision, Sim2O conceptualizes it as a compositional process. Specifically, candidate joint actions are synthesized by dynamically blending offline and online action proposals across agents. By leveraging a centralized value function to evaluate these hybrid combinations, Sim2O identifies high-value coordination strategies without requiring auxiliary training objectives or structural overhead. Empirical evaluations across diverse benchmarks demonstrate that Sim2O significantly outperforms existing baselines, underscoring that a minimalist design is not only viable but highly effective for multi-agent offline-to-online adaptation.

cs.LG↗

arXiv2026

COLLIE: Guiding Skill Discovery in Semantically Coherent Latent Space

Unsupervised skill discovery (USD) aims to learn diverse behaviors without reward functions, but often results in task-irrelevant or hazardous behaviors due to uniform exploration. Guided skill discovery (GSD) addresses this issue by incorporating human intent to focus exploration on meaningful regions. However, existing GSD methods typically require training additional guidance models, and rely on pre-defined rules or expert demonstration, which can be ineffective under sparse, online-collected human feedback. To overcome this, we propose COLLIE, a GSD framework that leverages dense unsupervised data to construct a semantically coherent skill latent space. This latent space is well-structured, enabling reliable guidance with sparse online feedback. Moreover, its semantic coherence property enables training-free construction of guidance signals, eliminating the need for additional model training beyond skill learning. Theoretical analysis justifies the effectiveness of our training-free guidance signal, while experiments across diverse state-based and pixel-based tasks show that COLLIE learns diverse, human-aligned skills, avoids hazardous behaviors, and achieves superior downstream performance with minimal human feedback.

cs.LG↗

arXiv2026

Curriculum reinforcement learning with measurable task representation learning

In curriculum reinforcement learning (CRL), an agent incrementally accumulates knowledge over a sequence of tasks (i.e., a curriculum), and the learning process is aimed at using the accumulated knowledge to finally solve a challenging target task. While early CRL works focus on sequencing candidate tasks, recent research explores automatic curriculum generation. Among the rich CRL literature, the interpolation-based CRL paradigm is a main body, which automatically generates intermediate tasks by interpolating between the initial task distribution and the target task distribution in task space with meaningful distance metrics (i.e., can measure the task similarity). However, in challenging navigation tasks, the non-Euclidean context (task) space invalidates this assumption. To achieve automatic curriculum generation in complex task, we propose a novel automatic curriculum generation approach based on measurable task representation learning. To better measure the similarity, we propose to transform the task space to a latent space. Through a variational autoencoder structure that encodes the reward and the state transitions, we achieve a latent task representation with a task similarity measurement property, and two close task embeddings correspond to two similar tasks in terms of rewards and state transitions. Based on the learned task representation, we further develop an automatic curriculum generation scheme, which can effectively generate new tasks more and more similar to the target task. We evaluate our method in a variety of challenging navigation tasks, and the experiment results indicate that the proposed approach surpasses state-of-the-art CRL approaches based on interpolation and generative adversarial networks.

cs.LG↗

arXiv2026

Data-Enabled Policy and Value Iteration for Continuous-Time Linear Quadratic Output Feedback Control

This paper proposes efficient policy iteration and value iteration algorithms for the continuous-time linear quadratic regulator problem with unmeasurable states and unknown system dynamics, from the perspective of direct data-driven control. Specifically, by re-examining the data characteristics of input-output filtered vectors and introducing QR decomposition, an improved substitute state construction method is presented that further eliminates redundant information, ensures a full row rank data matrix, and enables a complete parameterized representation of the feedback controller. Furthermore, the original problem is transformed into an equivalent linear quadratic regulator problem defined on the substitute state with a known input matrix, verifying the stabilizability and detectability of the transformed system. Consequently, model-free policy iteration and value iteration algorithms are designed that fully exploit the full row rank substitute state data matrix. The proposed algorithms offer distinct advantages: they avoid the need for prior knowledge of the system order or the calculation of signal derivatives and integrals; the iterative equations can be solved directly without relying on the traditional least-squares paradigm, guaranteeing feasibility in both single-output and multi-output settings; and they demonstrate superior numerical stability, reduced data demand, and higher computational efficiency. Moreover, the heuristic results regarding trajectory generation for continuous-time systems are discussed, circumventing potential failure modes associated with existing approaches.

eess.SY↗

arXiv2026

OPRIDE: Offline Preference-based Reinforcement Learning via In-Dataset Exploration

Preference-based reinforcement learning (PbRL) can help avoid sophisticated reward designs and align better with human intentions, showing great promise in various real-world applications. However, obtaining human feedback for preferences can be expensive and time-consuming, which forms a strong barrier for PbRL. In this work, we address the problem of low query efficiency in offline PbRL, pinpointing two primary reasons: inefficient exploration and overoptimization of learned reward functions. In response to these challenges, we propose a novel algorithm, \textbf{O}ffline \textbf{P}b\textbf{R}L via \textbf{I}n-\textbf{D}ataset \textbf{E}xploration (OPRIDE), designed to enhance the query efficiency of offline PbRL. OPRIDE consists of two key features: a principled exploration strategy that maximizes the informativeness of the queries and a discount scheduling mechanism aimed at mitigating overoptimization of the learned reward functions. Through empirical evaluations, we demonstrate that OPRIDE significantly outperforms prior methods, achieving strong performance with notably fewer queries. Moreover, we provide theoretical guarantees of the algorithm's efficiency. Experimental results across various locomotion, manipulation, and navigation tasks underscore the efficacy and versatility of our approach.

cs.LG↗

arXiv2026

GlobeDiff: State Diffusion Process for Partial Observability in Multi-Agent Systems

In the realm of multi-agent systems, the challenge of \emph{partial observability} is a critical barrier to effective coordination and decision-making. Existing approaches, such as belief state estimation and inter-agent communication, often fall short. Belief-based methods are limited by their focus on past experiences without fully leveraging global information, while communication methods often lack a robust model to effectively utilize the auxiliary information they provide. To solve this issue, we propose Global State Diffusion Algorithm~(GlobeDiff) to infer the global state based on the local observations. By formulating the state inference process as a multi-modal diffusion process, GlobeDiff overcomes ambiguities in state estimation while simultaneously inferring the global state with high fidelity. We prove that the estimation error of GlobeDiff under both unimodal and multi-modal distributions can be bounded. Extensive experimental results demonstrate that GlobeDiff achieves superior performance and is capable of accurately inferring the global state.

cs.AI↗

arXiv2025

MrCoM: A Meta-Regularized World-Model Generalizing Across Multi-Scenarios

Model-based reinforcement learning (MBRL) is a crucial approach to enhance the generalization capabilities and improve the sample efficiency of RL algorithms. However, current MBRL methods focus primarily on building world models for single tasks and rarely address generalization across different scenarios. Building on the insight that dynamics within the same simulation engine share inherent properties, we attempt to construct a unified world model capable of generalizing across different scenarios, named Meta-Regularized Contextual World-Model (MrCoM). This method first decomposes the latent state space into various components based on the dynamic characteristics, thereby enhancing the accuracy of world-model prediction. Further, MrCoM adopts meta-state regularization to extract unified representation of scenario-relevant information, and meta-value regularization to align world-model optimization with policy learning across diverse scenario objectives. We theoretically analyze the generalization error upper bound of MrCoM in multi-scenario settings. We systematically evaluate our algorithm's generalization ability across diverse scenarios, demonstrating significantly better performance than previous state-of-the-art methods.

cs.LG↗

arXiv2025

DAIL: Beyond Task Ambiguity for Language-Conditioned Reinforcement Learning

Comprehending natural language and following human instructions are critical capabilities for intelligent agents. However, the flexibility of linguistic instructions induces substantial ambiguity across language-conditioned tasks, severely degrading algorithmic performance. To address these limitations, we present a novel method named DAIL (Distributional Aligned Learning), featuring two key components: distributional policy and semantic alignment. Specifically, we provide theoretical results that the value distribution estimation mechanism enhances task differentiability. Meanwhile, the semantic alignment module captures the correspondence between trajectories and linguistic instructions. Extensive experimental results on both structured and visual observation benchmarks demonstrate that DAIL effectively resolves instruction ambiguities, achieving superior performance to baseline methods. Our implementation is available at https://github.com/RunpengXie/Distributional-Aligned-Learning.

cs.AI↗

arXiv2025

STAIR: Addressing Stage Misalignment through Temporal-Aligned Preference Reinforcement Learning

Preference-based reinforcement learning (PbRL) bypasses complex reward engineering by learning rewards directly from human preferences, enabling better alignment with human intentions. However, its effectiveness in multi-stage tasks, where agents sequentially perform sub-tasks (e.g., navigation, grasping), is limited by stage misalignment: Comparing segments from mismatched stages, such as movement versus manipulation, results in uninformative feedback, thus hindering policy learning. In this paper, we validate the stage misalignment issue through theoretical analysis and empirical experiments. To address this issue, we propose STage-AlIgned Reward learning (STAIR), which first learns a stage approximation based on temporal distance, then prioritizes comparisons within the same stage. Temporal distance is learned via contrastive learning, which groups temporally close states into coherent stages, without predefined task knowledge, and adapts dynamically to policy changes. Extensive experiments demonstrate STAIR's superiority in multi-stage tasks and competitive performance in single-stage tasks. Furthermore, human studies show that stages approximated by STAIR are consistent with human cognition, confirming its effectiveness in mitigating stage misalignment.

cs.LG↗

arXiv2025

An Efficient Data-Driven Framework for Linear Quadratic Output Feedback Control

Linear quadratic regulator with unmeasurable states and unknown system matrix parameters better aligns with practical scenarios. However, for this problem, balancing the optimality of the resulting controller and the leniency of the algorithm's feasibility conditions remains a non-trivial challenge, as no well-established general method has yet been developed to address this trade-off. To address this gap, this study first develops a comprehensive theoretical framework for state parameterization that equivalently substitutes for unknown states. By analyzing the controllability of consistent systems satisfied by substitute states, this framework quantifies the capability of substitute state data matrices to parameterize unknown closed-loop systems and output feedback controllers, thereby constructing a modified state parameterization form that meets the complete data parameterization condition of Willems' Fundamental Lemma. Leveraging this framework, this study proposes efficient model-free off-policy policy iteration and value iteration algorithms with theoretical guarantees to solve for the optimal output feedback controller. Compared with existing studies, particularly for multi-output problems where existing model-free reinforcement learning algorithms may fail, the proposed method removes redundant information in substitute states and the additional full row rank condition on regression matrices, thereby ensuring the solution of optimal output feedback controllers equivalent to optimal state feedback controllers for multi-output systems. Furthermore, this study pioneers a comprehensive and highly scalable theoretical analysis of state parameterization from a data-driven viewpoint, and the proposed algorithms exhibit significant advantages in implementation conditions, data demand, unknown handling, and convergence speed.

math.OC↗

arXiv2025

SC2Arena and StarEvolve: Benchmark and Self-Improvement Framework for LLMs in Complex Decision-Making Tasks

Evaluating large language models (LLMs) in complex decision-making is essential for advancing AI's ability for strategic planning and real-time adaptation. However, existing benchmarks for tasks like StarCraft II fail to capture the game's full complexity, such as its complete game context, diverse action spaces, and all playable races. To address this gap, we present SC2Arena, a benchmark that fully supports all playable races, low-level action spaces, and optimizes text-based observations to tackle spatial reasoning challenges. Complementing this, we introduce StarEvolve, a hierarchical framework that integrates strategic planning with tactical execution, featuring iterative self-correction and continuous improvement via fine-tuning on high-quality gameplay data. Its key components include a Planner-Executor-Verifier structure to break down gameplay, and a scoring system for selecting high-quality training samples. Comprehensive analysis using SC2Arena provides valuable insights into developing generalist agents that were not possible with previous benchmarks. Experimental results also demonstrate that our proposed StarEvolve achieves superior performance in strategic planning. Our code, environment, and algorithms are publicly available.

cs.LG↗

arXiv2025

DPMT: Dual Process Multi-scale Theory of Mind Framework for Real-time Human-AI Collaboration

Real-time human-artificial intelligence (AI) collaboration is crucial yet challenging, especially when AI agents must adapt to diverse and unseen human behaviors in dynamic scenarios. Existing large language model (LLM) agents often fail to accurately model the complex human mental characteristics such as domain intentions, especially in the absence of direct communication. To address this limitation, we propose a novel dual process multi-scale theory of mind (DPMT) framework, drawing inspiration from cognitive science dual process theory. Our DPMT framework incorporates a multi-scale theory of mind (ToM) module to facilitate robust human partner modeling through mental characteristic reasoning. Experimental results demonstrate that DPMT significantly enhances human-AI collaboration, and ablation studies further validate the contributions of our multi-scale ToM in the slow system.

cs.LG↗

arXiv2025

CLARIFY: Contrastive Preference Reinforcement Learning for Untangling Ambiguous Queries

Preference-based reinforcement learning (PbRL) bypasses explicit reward engineering by inferring reward functions from human preference comparisons, enabling better alignment with human intentions. However, humans often struggle to label a clear preference between similar segments, reducing label efficiency and limiting PbRL's real-world applicability. To address this, we propose an offline PbRL method: Contrastive LeArning for ResolvIng Ambiguous Feedback (CLARIFY), which learns a trajectory embedding space that incorporates preference information, ensuring clearly distinguished segments are spaced apart, thus facilitating the selection of more unambiguous queries. Extensive experiments demonstrate that CLARIFY outperforms baselines in both non-ideal teachers and real human feedback settings. Our approach not only selects more distinguished queries but also learns meaningful trajectory embeddings.

cs.LG↗

arXiv2025

S-EPOA: Overcoming the Indistinguishability of Segments with Skill-Driven Preference-Based Reinforcement Learning

Preference-based reinforcement learning (PbRL) stands out by utilizing human preferences as a direct reward signal, eliminating the need for intricate reward engineering. However, despite its potential, traditional PbRL methods are often constrained by the indistinguishability of segments, which impedes the learning process. In this paper, we introduce Skill-Enhanced Preference Optimization Algorithm (S-EPOA), which addresses the segment indistinguishability issue by integrating skill mechanisms into the preference learning framework. Specifically, we first conduct the unsupervised pretraining to learn useful skills. Then, we propose a novel query selection mechanism to balance the information gain and distinguishability over the learned skill space. Experimental results on a range of tasks, including robotic manipulation and locomotion, demonstrate that S-EPOA significantly outperforms conventional PbRL methods in terms of both robustness and learning efficiency. The results highlight the effectiveness of skill-driven learning in overcoming the challenges posed by segment indistinguishability.

cs.AI↗

arXiv2025

Fewer May Be Better: Enhancing Offline Reinforcement Learning with Reduced Dataset

Offline reinforcement learning (RL) represents a significant shift in RL research, allowing agents to learn from pre-collected datasets without further interaction with the environment. A key, yet underexplored, challenge in offline RL is selecting an optimal subset of the offline dataset that enhances both algorithm performance and training efficiency. Reducing dataset size can also reveal the minimal data requirements necessary for solving similar problems. In response to this challenge, we introduce ReDOR (Reduced Datasets for Offline RL), a method that frames dataset selection as a gradient approximation optimization problem. We demonstrate that the widely used actor-critic framework in RL can be reformulated as a submodular optimization objective, enabling efficient subset selection. To achieve this, we adapt orthogonal matching pursuit (OMP), incorporating several novel modifications tailored for offline RL. Our experimental results show that the data subsets identified by ReDOR not only boost algorithm performance but also do so with significantly lower computational complexity.

cs.LG↗

arXiv2025

Episodic Novelty Through Temporal Distance

Exploration in sparse reward environments remains a significant challenge in reinforcement learning, particularly in Contextual Markov Decision Processes (CMDPs), where environments differ across episodes. Existing episodic intrinsic motivation methods for CMDPs primarily rely on count-based approaches, which are ineffective in large state spaces, or on similarity-based methods that lack appropriate metrics for state comparison. To address these shortcomings, we propose Episodic Novelty Through Temporal Distance (ETD), a novel approach that introduces temporal distance as a robust metric for state similarity and intrinsic reward computation. By employing contrastive learning, ETD accurately estimates temporal distances and derives intrinsic rewards based on the novelty of states within the current episode. Extensive experiments on various benchmark tasks demonstrate that ETD significantly outperforms state-of-the-art methods, highlighting its effectiveness in enhancing exploration in sparse reward CMDPs.

cs.LG↗