Jialin Liu

Qingquan Zhang, Jialin Liu, Zeqi Zhang et al.

In the literature of mitigating unfairness in machine learning, many fairness measures are designed to evaluate predictions of learning models and also utilised to guide the training of fair models. It has been theoretically and empirically shown that there exist conflicts and inconsistencies among accuracy and multiple fairness measures. Optimising one or several fairness measures may sacrifice or deteriorate other measures. Two key questions should be considered, how to simultaneously optimise accuracy and multiple fairness measures, and how to optimise all the considered fairness measures more effectively. In this paper, we view the mitigating unfairness problem as a multi-objective learning problem considering the conflicts among fairness measures. A multi-objective evolutionary learning framework is used to simultaneously optimise several metrics (including accuracy and multiple fairness measures) of machine learning models. Then, ensembles are constructed based on the learning models in order to automatically balance different metrics. Empirical results on eight well-known datasets demonstrate that compared with the state-of-the-art approaches for mitigating unfairness, our proposed algorithm can provide decision-makers with better tradeoffs among accuracy and multiple fairness metrics. Furthermore, the high-quality models generated by the framework can be used to construct an ensemble to automatically achieve a better tradeoff among all the considered fairness metrics than other ensemble methods. Our code is publicly available at https://github.com/qingquan63/FairEMOL

Chengpeng Hu, Yunlong Zhao, Ziqi Wang et al.

Games have been the perfect test-beds for artificial intelligence research for the characteristics that widely exist in real-world scenarios. Learning and optimisation, decision making in dynamic and uncertain environments, game theory, planning and scheduling, design and education are common research areas shared between games and real-world problems. Numerous open-source games or game-based environments have been implemented for studying artificial intelligence. In addition to single- or multi-player, collaborative or adversarial games, there has also been growing interest in implementing platforms for creative design in recent years. Those platforms provide ideal benchmarks for exploring and comparing artificial intelligence ideas and techniques. This paper reviews the games and game-based platforms for artificial intelligence research, provides guidance on matching particular types of artificial intelligence with suitable games for testing and matching particular needs in games with suitable artificial intelligence techniques, discusses the research trend induced by the evolution of those games and platforms, and gives an outlook.

Ziang Chen, Jialin Liu, Xinshang Wang et al.

Learning to optimize is a rapidly growing area that aims to solve optimization problems or improve existing optimization algorithms using machine learning (ML). In particular, the graph neural network (GNN) is considered a suitable ML model for optimization problems whose variables and constraints are permutation--invariant, for example, the linear program (LP). While the literature has reported encouraging numerical results, this paper establishes the theoretical foundation of applying GNNs to solving LPs. Given any size limit of LPs, we construct a GNN that maps different LPs to different outputs. We show that properly built GNNs can reliably predict feasibility, boundedness, and an optimal solution for each LP in a broad class. Our proofs are based upon the recently--discovered connections between the Weisfeiler--Lehman isomorphism test and the GNN. To validate our results, we train a simple GNN and present its accuracy in mapping LPs to their feasibilities and solutions.

Ziang Chen, Jialin Liu, Xinshang Wang et al.

While Mixed-integer linear programming (MILP) is NP-hard in general, practical MILP has received roughly 100--fold speedup in the past twenty years. Still, many classes of MILPs quickly become unsolvable as their sizes increase, motivating researchers to seek new acceleration techniques for MILPs. With deep learning, they have obtained strong empirical results, and many results were obtained by applying graph neural networks (GNNs) to making decisions in various stages of MILP solution processes. This work discovers a fundamental limitation: there exist feasible and infeasible MILPs that all GNNs will, however, treat equally, indicating GNN's lacking power to express general MILPs. Then, we show that, by restricting the MILPs to unfoldable ones or by adding random features, there exist GNNs that can reliably predict MILP feasibility, optimal objective values, and optimal solutions up to prescribed precision. We conducted small-scale numerical experiments to validate our theoretical findings.

InSpatio Team, Xiaoyu Zhang, Weihong Pan et al.

We present InSpatio-WorldFM, an open-source real-time frame model for spatial intelligence. Unlike video-based world models that rely on sequential frame generation and incur substantial latency due to window-level processing, InSpatio-WorldFM adopts a frame-based paradigm that generates each frame independently, enabling low-latency real-time spatial inference. By enforcing multi-view spatial consistency through explicit 3D anchors and implicit spatial memory, the model preserves global scene geometry while maintaining fine-grained visual details across viewpoint changes. We further introduce a progressive three-stage training pipeline that transforms a pretrained image diffusion model into a controllable frame model and finally into a real-time generator through few-step distillation. Experimental results show that InSpatio-WorldFM achieves strong multi-view consistency while supporting interactive exploration on consumer-grade GPUs, providing an efficient alternative to traditional video-based world models for real-time world simulation.

Shuo Huang, Chengpeng Hu, Julian Togelius et al.

Procedurally generating cities in Minecraft provides players more diverse scenarios and could help understand and improve the design of cities in other digital worlds and the real world. This paper presents a city generator that was submitted as an entry to the 2023 Edition of Minecraft Settlement Generation Competition for Minecraft. The generation procedure is composed of six main steps, namely vegetation clearing, terrain reshaping, building layout generation, route planning, streetlight placement, and wall construction. Three algorithms, including a heuristic-based algorithm, an evolving layout algorithm, and a random one are applied to generate the building layout, thus determining where to place different redstone style buildings, and tested by generating cities on random maps in limited time. Experimental results show that the heuristic-based algorithm is capable of finding an acceptable building layout faster for flat maps, while the evolving layout algorithm performs better in evolving layout for rugged maps. A user study is conducted to compare our generator with outstanding entries of the competition's 2022 edition using the competition's evaluation criteria and shows that our generator performs well in the adaptation and functionality criteria

Keyuan Zhang, Jiayu Bai, Jialin Liu

Recent years, there has been growing interests in experience-driven procedural level generation. Various metrics have been formulated to model player experience and help generate personalised levels. In this work, we question whether experience metrics can adapt to agents with different personas. We start by reviewing existing metrics for evaluating game levels. Then, focusing on platformer games, we design a framework integrating various agents and evaluation metrics. Experimental studies on \emph{Super Mario Bros.} indicate that using the same evaluation metrics but agents with different personas can generate levels for particular persona. It implies that, for simple games, using a game-playing agent of specific player archetype as a level tester is probably all we need to generate levels of diverse behaviour engagement.

Ge Fan, Nan Zhao, Kai Meng et al.

With the rapid evolution of internet services, recommendation systems have become indispensable. In particular, the blending (re-ranking) stage plays a pivotal role in allocating traffic across diverse business objectives. However, existing approaches often suffer from coupled allocation plans, score inflation, and a lack of interpretability. To address these challenges, we propose Uniboost, a unified traffic allocation framework. Uniboost introduces a posterior value alignment mechanism that calibrates abstract model scores to anchor metrics with explicit business semantics, significantly enhancing interpretability. Furthermore, it employs an independent linear boosting paradigm to decouple complex weighting schemes, enabling precise attribution of each plan's contribution. We validate the effectiveness of Uniboost through online A/B tests and in-depth data analysis, demonstrating three key findings: 1) Reducing the overall weight of weighted scores effectively mitigates unintended business interference, yielding a more efficient micro-level traffic allocation strategy; 2) Post-hoc analyses and aggregated dashboards provide intuitive, macro-level insights that guide the design of the overall traffic allocation mechanism; 3) The proposed "Effective Completion Score" serves as an easily obtainable post-metric that offers a reliable anchor for content recommendation pipelines. Collectively, our experiments show that Uniboost not only improves traffic allocation efficiency and recommendation performance at the micro level but also provides macro-level guidance for system iteration. Thus, this work provides an efficient and controllable traffic regulation solution for large-scale industrial recommendation systems.

Jialin Liu, Ning Miao, Chongzhou Fang et al.

The Electrocardiogram (ECG) measures the electrical cardiac activity generated by the heart to detect abnormal heartbeat and heart attack. However, the irregular occurrence of the abnormalities demands continuous monitoring of heartbeats. Machine learning techniques are leveraged to automate the task to reduce labor work needed during monitoring. In recent years, many companies have launched products with ECG monitoring and irregular heartbeat alert. Among all classification algorithms, the time series-based algorithm dynamic time warping (DTW) is widely adopted to undertake the ECG classification task. Though progress has been achieved, the DTW-based ECG classification also brings a new attacking vector of leaking the patients' diagnosis results. This paper shows that the ECG input samples' labels can be stolen via a side-channel attack, Flush+Reload. In particular, we first identify the vulnerability of DTW for ECG classification, i.e., the correlation between warping path choice and prediction results. Then we implement an attack that leverages Flush+Reload to monitor the warping path selection with known ECG data and then build a predictor for constructing the relation between warping path selection and labels of input ECG samples. Based on experiments, we find that the Flush+Reload-based inference leakage can achieve an 84.0\% attacking success rate to identify the labels of the two samples in DTW.

Ziqi Wang, Jialin Liu

Game consists of multiple types of content, while the harmony of different content types play an essential role in game design. However, most works on procedural content generation consider only one type of content at a time. In this paper, we propose and formulate online level generation from music, in a way of matching a level feature to a music feature in real-time, while adapting to players' play speed. A generic framework named online player-adaptive procedural content generation via reinforcement learning, OPARL for short, is built upon the experience-driven reinforcement learning and controllable reinforcement learning, to enable online level generation from music. Furthermore, a novel control policy based on local search and k-nearest neighbours is proposed and integrated into OPARL to control the level generator considering the play data collected online. Results of simulation-based experiments show that our implementation of OPARL is competent to generate playable levels with difficulty degree matched to the ``energy'' dynamic of music for different artificial players in an online fashion.

Haoyu Wang, Jialin Liu, Xiaohan Chen et al.

Mixed-integer linear programming (MILP) stands as a notable NP-hard problem pivotal to numerous crucial industrial applications. The development of effective algorithms, the tuning of solvers, and the training of machine learning models for MILP resolution all hinge on access to extensive, diverse, and representative data. Yet compared to the abundant naturally occurring data in image and text realms, MILP is markedly data deficient, underscoring the vital role of synthetic MILP generation. We present DIG-MILP, a deep generative framework based on variational auto-encoder (VAE), adept at extracting deep-level structural features from highly limited MILP data and producing instances that closely mirror the target data. Notably, by leveraging the MILP duality, DIG-MILP guarantees a correct and complete generation space as well as ensures the boundedness and feasibility of the generated instances. Our empirical study highlights the novelty and quality of the instances generated by DIG-MILP through two distinct downstream tasks: (S1) Data sharing, where solver solution times correlate highly positive between original and DIG-MILP-generated instances, allowing data sharing for solver tuning without publishing the original data; (S2) Data Augmentation, wherein the DIG-MILP-generated instances bolster the generalization performance of machine learning models tasked with resolving MILP problems.

Chengpeng Hu, Jiyuan Pei, Jialin Liu et al.

Evolutionary algorithms have been used to evolve a population of actors to generate diverse experiences for training reinforcement learning agents, which helps to tackle the temporal credit assignment problem and improves the exploration efficiency. However, when adapting this approach to address constrained problems, balancing the trade-off between the reward and constraint violation is hard. In this paper, we propose a novel evolutionary constrained reinforcement learning (ECRL) algorithm, which adaptively balances the reward and constraint violation with stochastic ranking, and at the same time, restricts the policy's behaviour by maintaining a set of Lagrange relaxation coefficients with a constraint buffer. Extensive experiments on robotic control benchmarks show that our ECRL achieves outstanding performance compared to state-of-the-art algorithms. Ablation analysis shows the benefits of introducing stochastic ranking and constraint buffer.

Shaobo Wang, Xuan Ouyang, Tianyi Xu et al.

As high-quality public text approaches exhaustion, a phenomenon known as the Data Wall, pre-training is shifting from more tokens to better tokens. However, existing methods either rely on heuristic static filters that ignore training dynamics, or use dynamic yet optimizer-agnostic criteria based on raw gradients. We propose OPUS (Optimizer-induced Projected Utility Selection), a dynamic data selection framework that defines utility in the optimizer-induced update space. OPUS scores candidates by projecting their effective updates, shaped by modern optimizers, onto a target direction derived from a stable, in-distribution proxy. To ensure scalability, we employ Ghost technique with CountSketch for computational efficiency, and Boltzmann sampling for data diversity, incurring only 4.7\% additional compute overhead. OPUS achieves remarkable results across diverse corpora, quality tiers, optimizers, and model scales. In pre-training of GPT-2 Large/XL on FineWeb and FineWeb-Edu with 30B tokens, OPUS outperforms industrial-level baselines and even full 200B-token training. Moreover, when combined with industrial-level static filters, OPUS further improves pre-training efficiency, even with lower-quality data. Furthermore, in continued pre-training of Qwen3-8B-Base on SciencePedia, OPUS achieves superior performance using only 0.5B tokens compared to full training with 3B tokens, demonstrating significant data efficiency gains in specialized domains.

Qingquan Zhang, Jialin Liu, Xin Yao

Multiobjective evolutionary learning (MOEL) has demonstrated its advantages of training fairer machine learning models considering a predefined set of conflicting objectives, including accuracy and different fairness measures. Recent works propose to construct a representative subset of fairness measures as optimisation objectives of MOEL throughout model training. However, the determination of a representative measure set relies on dataset, prior knowledge and requires substantial computational costs. What's more, those representative measures may differ across different model training processes. Instead of using a static predefined set determined before model training, this paper proposes to dynamically and adaptively determine a representative measure set online during model training. The dynamically determined representative set is then used as optimising objectives of the MOEL framework and can vary with time. Extensive experimental results on 12 well-known benchmark datasets demonstrate that our proposed framework achieves outstanding performance compared to state-of-the-art approaches for mitigating unfairness in terms of accuracy as well as 25 fairness measures although only a few of them were dynamically selected and used as optimisation objectives. The results indicate the importance of setting optimisation objectives dynamically during training.

Xue Wang, Tian Zhou, Jianqing Zhu et al.

Attention based models have achieved many remarkable breakthroughs in numerous applications. However, the quadratic complexity of Attention makes the vanilla Attention based models hard to apply to long sequence tasks. Various improved Attention structures are proposed to reduce the computation cost by inducing low rankness and approximating the whole sequence by sub-sequences. The most challenging part of those approaches is maintaining the proper balance between information preservation and computation reduction: the longer sub-sequences used, the better information is preserved, but at the price of introducing more noise and computational costs. In this paper, we propose a smoothed skeleton sketching based Attention structure, coined S$^3$Attention, which significantly improves upon the previous attempts to negotiate this trade-off. S$^3$Attention has two mechanisms to effectively minimize the impact of noise while keeping the linear complexity to the sequence length: a smoothing block to mix information over long sequences and a matrix sketching method that simultaneously selects columns and rows from the input matrix. We verify the effectiveness of S$^3$Attention both theoretically and empirically. Extensive studies over Long Range Arena (LRA) datasets and six time-series forecasting show that S$^3$Attention significantly outperforms both vanilla Attention and other state-of-the-art variants of Attention structures.

Jiyuan Pei, Yi Mei, Jialin Liu et al.

Existing neural solvers for vehicle routing problems (VRPs) are typically trained either in a one-off manner on a fixed set of pre-defined tasks or in a lifelong manner on several tasks arriving sequentially, assuming sufficient training on each task. Both settings overlook a common real-world property: problem patterns may drift continually over time, yielding massive tasks sequentially arising while offering only limited training resources per task. In this paper, we study a novel lifelong learning paradigm for neural VRP solvers under continually drifting tasks over learning time steps, where sufficient training for any given task at any time is not available. We propose Dual Replay with Experience Enhancement (DREE), a general framework to improve learning efficiency and mitigate catastrophic forgetting under such drift. Extensive experiments show that, under such continual drift, DREE effectively learns new tasks, preserves prior knowledge, improves generalization to unseen tasks, and can be applied to diverse existing neural solvers.

Ziqi Wang, Tianye Shu, Jialin Liu

In this paper, we revisit endless online level generation with the recently proposed experience-driven procedural content generation via reinforcement learning (EDRL) framework. Inspired by an observation that EDRL tends to generate recurrent patterns, we formulate a notion of state space closure which makes any stochastic state appeared possibly in an infinite-horizon online generation process can be found within a finite-horizon. Through theoretical analysis, we find that even though state space closure arises a concern about diversity, it generalises EDRL trained with a finite-horizon to the infinite-horizon scenario without deterioration of content quality. Moreover, we verify the quality and the diversity of contents generated by EDRL via empirical studies, on the widely used Super Mario Bros. benchmark. Experimental results reveal that the diversity of levels generated by EDRL is limited due to the state space closure, whereas their quality does not deteriorate in a horizon which is longer than the one specified in the training. Concluding our outcomes and analysis, future work on endless online level generation via reinforcement learning should address the issue of diversity while assuring the occurrence of state space closure and quality.

Jialin Liu, Lu Yan, Xiaowei Liu et al.

n clinical, if a patient presents with nonmechanical obstructive dysphagia, esophageal chest pain, and gastro esophageal reflux symptoms, the physician will usually assess the esophageal dynamic function. High-resolution manometry (HRM) is a clinically commonly used technique for detection of esophageal dynamic function comprehensively and objectively. However, after the results of HRM are obtained, doctors still need to evaluate by a variety of parameters. This work is burdensome, and the process is complex. We conducted image processing of HRM to predict the esophageal contraction vigor for assisting the evaluation of esophageal dynamic function. Firstly, we used Feature-Extraction and Histogram of Gradients (FE-HOG) to analyses feature of proposal of swallow (PoS) to further extract higher-order features. Then we determine the classification of esophageal contraction vigor normal, weak and failed by using linear-SVM according to these features. Our data set includes 3000 training sets, 500 validation sets and 411 test sets. After verification our accuracy reaches 86.83%, which is higher than other common machine learning methods.

Jialin Liu, Jianhua Wu, Jie Liu et al.

Fine-tuning large language models (LLMs) with Low-Rank adaption (LoRA) is widely acknowledged as an effective approach for continual learning for new tasks. However, it often suffers from catastrophic forgetting when dealing with multiple tasks sequentially. To this end, we propose Attentional Mixture of LoRAs (AM-LoRA), a continual learning approach tailored for LLMs. Specifically, AM-LoRA learns a sequence of LoRAs for a series of tasks to continually learn knowledge from different tasks. The key of our approach is that we devise an attention mechanism as a knowledge mixture module to adaptively integrate information from each LoRA. With the attention mechanism, AM-LoRA can efficiently leverage the distinctive contributions of each LoRA, while mitigating the risk of mutually negative interactions among them that may lead to catastrophic forgetting. Moreover, we further introduce $L1$ norm in the learning process to make the attention vector more sparse. The sparse constraints can enable the model to lean towards selecting a few highly relevant LoRAs, rather than aggregating and weighting all LoRAs collectively, which can further reduce the impact stemming from mutual interference. Experimental results on continual learning benchmarks indicate the superiority of our proposed method.

Meituan LongCat Team, Anchun Gui, Bei Li et al.

We present LongCat-Flash-Thinking, an efficient 560-billion-parameter open-source Mixture-of-Experts (MoE) reasoning model. Its advanced capabilities are cultivated through a meticulously crafted training process, beginning with long Chain-of-Thought (CoT) data cold-start and culminating in large-scale Reinforcement Learning (RL). We first employ a well-designed cold-start training strategy, which significantly enhances the reasoning potential and equips the model with specialized skills in both formal and agentic reasoning. Then, a core innovation is our domain-parallel training scheme, which decouples optimization across distinct domains (e.g., STEM, Code, Agentic) and subsequently fuses the resulting expert models into a single, nearly Pareto-optimal model. This entire process is powered by our Dynamic ORchestration for Asynchronous rollout (DORA) system, a large-scale RL framework that delivers a greater than threefold training speedup over synchronous methods on tens of thousands of accelerators. As a result, LongCat-Flash-Thinking achieves state-of-the-art performance among open-source models on a suite of complex reasoning tasks. The model exhibits exceptional efficiency in agentic reasoning, reducing average token consumption by 64.5% (from 19, 653 to 6, 965) on AIME-25, without degrading task accuracy. We release LongCat-Flash-Thinking to promote further advances in reasoning systems and agentic AI research.

Xiaohan Chen, Jialin Liu, Zhangyang Wang et al.

Learned Iterative Shrinkage-Thresholding Algorithm (LISTA) introduces the concept of unrolling an iterative algorithm and training it like a neural network. It has had great success on sparse recovery. In this paper, we show that adding momentum to intermediate variables in the LISTA network achieves a better convergence rate and, in particular, the network with instance-optimal parameters is superlinearly convergent. Moreover, our new theoretical results lead to a practical approach of automatically and adaptively calculating the parameters of a LISTA network layer based on its previous layers. Perhaps most surprisingly, such an adaptive-parameter procedure reduces the training of LISTA to tuning only three hyperparameters from data: a new record set in the context of the recent advances on trimming down LISTA complexity. We call this new ultra-light weight network HyperLISTA. Compared to state-of-the-art LISTA models, HyperLISTA achieves almost the same performance on seen data distributions and performs better when tested on unseen distributions (specifically, those with different sparsity levels and nonzero magnitudes). Code is available: https://github.com/VITA-Group/HyperLISTA.

Tianlong Chen, Xiaohan Chen, Wuyang Chen et al.

Learning to optimize (L2O) is an emerging approach that leverages machine learning to develop optimization methods, aiming at reducing the laborious iterations of hand engineering. It automates the design of an optimization method based on its performance on a set of training problems. This data-driven procedure generates methods that can efficiently solve problems similar to those in the training. In sharp contrast, the typical and traditional designs of optimization methods are theory-driven, so they obtain performance guarantees over the classes of problems specified by the theory. The difference makes L2O suitable for repeatedly solving a certain type of optimization problems over a specific distribution of data, while it typically fails on out-of-distribution problems. The practicality of L2O depends on the type of target optimization, the chosen architecture of the method to learn, and the training procedure. This new paradigm has motivated a community of researchers to explore L2O and report their findings. This article is poised to be the first comprehensive survey and benchmark of L2O for continuous optimization. We set up taxonomies, categorize existing works and research directions, present insights, and identify open challenges. We also benchmarked many existing L2O approaches on a few but representative optimization problems. For reproducible research and fair benchmarking purposes, we released our software implementation and data in the package Open-L2O at https://github.com/VITA-Group/Open-L2O.

Jialin Liu, Fei Chao, Chih-Min Lin

Data augmentation is one of the most effective approaches for improving the accuracy of modern machine learning models, and it is also indispensable to train a deep model for meta-learning. In this paper, we introduce a task augmentation method by rotating, which increases the number of classes by rotating the original images 90, 180 and 270 degrees, different from traditional augmentation methods which increase the number of images. With a larger amount of classes, we can sample more diverse task instances during training. Therefore, task augmentation by rotating allows us to train a deep network by meta-learning methods with little over-fitting. Experimental results show that our approach is better than the rotation for increasing the number of images and achieves state-of-the-art performance on miniImageNet, CIFAR-FS, and FC100 few-shot learning benchmarks. The code is available on \url{www.github.com/AceChuse/TaskLevelAug}.

Xiaohan Chen, Jialin Liu, Zhangyang Wang et al.

In recent years, unfolding iterative algorithms as neural networks has become an empirical success in solving sparse recovery problems. However, its theoretical understanding is still immature, which prevents us from fully utilizing the power of neural networks. In this work, we study unfolded ISTA (Iterative Shrinkage Thresholding Algorithm) for sparse signal recovery. We introduce a weight structure that is necessary for asymptotic convergence to the true sparse signal. With this structure, unfolded ISTA can attain a linear convergence, which is better than the sublinear convergence of ISTA/FISTA in general cases. Furthermore, we propose to incorporate thresholding in the network to perform support selection, which is easy to implement and able to boost the convergence rate both theoretically and empirically. Extensive simulations, including sparse vector recovery and a compressive sensing experiment on real image data, corroborate our theoretical results and demonstrate their practical usefulness. We have made our codes publicly available: https://github.com/xchen-tamu/linear-lista-cpss.

Diego Perez-Liebana, Jialin Liu, Ahmed Khalifa et al.

General Video Game Playing (GVGP) aims at designing an agent that is capable of playing multiple video games with no human intervention. In 2014, The General Video Game AI (GVGAI) competition framework was created and released with the purpose of providing researchers a common open-source and easy to use platform for testing their AI methods with potentially infinity of games created using Video Game Description Language (VGDL). The framework has been expanded into several tracks during the last few years to meet the demand of different research directions. The agents are required either to play multiple unknown games with or without access to game simulations, or to design new game levels or rules. This survey paper presents the VGDL, the GVGAI framework, existing tracks, and reviews the wide use of GVGAI framework in research, education and competitions five years after its birth. A future plan of framework improvements is also described.

Caleb Jore, Jialin Liu

Many scientific and combinatorial problems admit multiple correct solutions, not a single label. Standard supervised learning resolves this ambiguity by choosing one solution as the target, but this hidden selector can be arbitrary, discontinuous, and harder to learn than the underlying solution set. We study bifurcation models, a weight-tied dynamical view in which different initializations can converge to different stable equilibria, so the model represents an attractor landscape rather than one chosen branch. We prove that broad set-valued maps with locally Lipschitz branches can be represented by regular equilibrium dynamics and that the induced selectors are almost everywhere regular, while manual selectors can be arbitrarily irregular. Experiments on frustrated Ising models show that such dynamics can discover multiple valid equilibria without branch labels and outperform single-branch supervision. Allen--Cahn experiments further show that diversity is not automatic: it can be encouraged explicitly, but with an accuracy--diversity tradeoff.

Xinda Wang, Zhengxu Hou, Yangshijie Zhang et al.

Reinforcement learning for open-ended text generation is constrained by the lack of verifiable rewards, necessitating reliance on judge models that require either annotated data or powerful closed-source models. Inspired by recent work on unsupervised reinforcement learning for mathematical reasoning using confidence-based endogenous rewards, we investigate whether this principle can be adapted to open-ended writing tasks. We find that directly applying confidence rewards leads to Triviality Bias: the policy collapses toward high-probability outputs, reducing diversity and meaningful content. We propose TCER (Triviality Corrected Endogenous Reward), which addresses this bias by rewarding the relative information gain between a specialist policy and a generalist reference policy, modulated by a probability-dependent correction mechanism. Across multiple writing benchmarks and model architectures, TCER achieves consistent improvements without external supervision. Furthermore, TCER also transfers effectively to mathematical reasoning, validating the generality of our approach across different generation tasks.

Jialin Liu, Zhaorui Zhang, Ray C. C. Cheung

Graph neural networks (GNNs) have revolutionized recommender systems by effectively modeling complex user-item interactions, yet data sparsity and the item cold-start problem significantly impair performance, particularly for new items with limited or no interaction history. While multimodal content offers a promising solution, existing methods result in suboptimal representations for new items due to noise and entanglement in sparse data. To address this, we transform multimodal recommendation into discrete semantic tokenization. We present Sparse-Regularized Multimodal Tokenization for Cold-Start Recommendation (MoToRec), a framework centered on a sparsely-regularized Residual Quantized Variational Autoencoder (RQ-VAE) that generates a compositional semantic code of discrete, interpretable tokens, promoting disentangled representations. MoToRec's architecture is enhanced by three synergistic components: (1) a sparsely-regularized RQ-VAE that promotes disentangled representations, (2) a novel adaptive rarity amplification that promotes prioritized learning for cold-start items, and (3) a hierarchical multi-source graph encoder for robust signal fusion with collaborative signals. Extensive experiments on three large-scale datasets demonstrate MoToRec's superiority over state-of-the-art methods in both overall and cold-start scenarios. Our work validates that discrete tokenization provides an effective and scalable alternative for mitigating the long-standing cold-start challenge.

Chengpeng Hu, Yunlong Zhao, Jialin Liu

Recently, the emergence of large language models (LLMs) has unlocked new opportunities for procedural content generation. However, recent attempts mainly focus on level generation for specific games with defined game rules such as Super Mario Bros. and Zelda. This paper investigates the game generation via LLMs. Based on video game description language, this paper proposes an LLM-based framework to generate game rules and levels simultaneously. Experiments demonstrate how the framework works with prompts considering different combinations of context. Our findings extend the current applications of LLMs and offer new insights for generating new games in the area of procedural content generation.

Jialin Liu, Xinyan Su, Zeyu He et al.

Rewards serve as a measure of user satisfaction and act as a limiting factor in interactive recommender systems. In this research, we focus on the problem of learning to reward (LTR), which is fundamental to reinforcement learning. Previous approaches either introduce additional procedures for learning to reward, thereby increasing the complexity of optimization, or assume that user-agent interactions provide perfect demonstrations, which is not feasible in practice. Ideally, we aim to employ a unified approach that optimizes both the reward and policy using compositional demonstrations. However, this requirement presents a challenge since rewards inherently quantify user feedback on-policy, while recommender agents approximate off-policy future cumulative valuation. To tackle this challenge, we propose a novel batch inverse reinforcement learning paradigm that achieves the desired properties. Our method utilizes discounted stationary distribution correction to combine LTR and recommender agent evaluation. To fulfill the compositional requirement, we incorporate the concept of pessimism through conservation. Specifically, we modify the vanilla correction using Bellman transformation and enforce KL regularization to constrain consecutive policy updates. We use two real-world datasets which represent two compositional coverage to conduct empirical studies, the results also show that the proposed method relatively improves both effectiveness (2.3\%) and efficiency (11.53\%)

Jialin Liu, Xinyan Su, Peng Zhou et al.

Survivor bias in observational data leads the optimization of recommender systems towards local optima. Currently most solutions re-mines existing human-system collaboration patterns to maximize longer-term satisfaction by reinforcement learning. However, from the causal perspective, mitigating survivor effects requires answering a counterfactual problem, which is generally unidentifiable and inestimable. In this work, we propose a neural causal model to achieve counterfactual inference. Specifically, we first build a learnable structural causal model based on its available graphical representations which qualitatively characterizes the preference transitions. Mitigation of the survivor bias is achieved though counterfactual consistency. To identify the consistency, we use the Gumbel-max function as structural constrains. To estimate the consistency, we apply reinforcement optimizations, and use Gumbel-Softmax as a trade-off to get a differentiable function. Both theoretical and empirical studies demonstrate the effectiveness of our solution.

Xiaohan Chen, Jialin Liu, Wotao Yin

Learning to Optimize (L2O) stands at the intersection of traditional optimization and machine learning, utilizing the capabilities of machine learning to enhance conventional optimization techniques. As real-world optimization problems frequently share common structures, L2O provides a tool to exploit these structures for better or faster solutions. This tutorial dives deep into L2O techniques, introducing how to accelerate optimization algorithms, promptly estimate the solutions, or even reshape the optimization problem itself, making it more adaptive to real-world applications. By considering the prerequisites for successful applications of L2O and the structure of the optimization problems at hand, this tutorial provides a comprehensive guide for practitioners and researchers alike.

InSpatio Team, Donghui Shen, Guofeng Zhang et al.

Building world models with spatial consistency and real-time interactivity remains a fundamental challenge in computer vision. Current video generation paradigms often struggle with a lack of spatial persistence and insufficient visual realism, making it difficult to support seamless navigation in complex environments. To address these challenges, we propose INSPATIO-WORLD, a novel real-time framework capable of recovering and generating high-fidelity, dynamic interactive scenes from a single reference video. At the core of our approach is a Spatiotemporal Autoregressive (STAR) architecture, which enables consistent and controllable scene evolution through two tightly coupled components: Implicit Spatiotemporal Cache aggregates reference and historical observations into a latent world representation, ensuring global consistency during long-horizon navigation; Explicit Spatial Constraint Module enforces geometric structure and translates user interactions into precise and physically plausible camera trajectories. Furthermore, we introduce Joint Distribution Matching Distillation (JDMD). By using real-world data distributions as a regularizing guide, JDMD effectively overcomes the fidelity degradation typically caused by over-reliance on synthetic data. Extensive experiments demonstrate that INSPATIO-WORLD significantly outperforms existing state-of-the-art (SOTA) models in spatial consistency and interaction precision, ranking first among real-time interactive methods on the WorldScore-Dynamic benchmark, and establishing a practical pipeline for navigating 4D environments reconstructed from monocular videos.

Ziang Chen, Jialin Liu, Xiaohan Chen et al.

Graph neural networks (GNNs) have been widely used to predict properties and heuristics of mixed-integer linear programs (MILPs) and hence accelerate MILP solvers. This paper investigates the capacity of GNNs to represent strong branching (SB), the most effective yet computationally expensive heuristic employed in the branch-and-bound algorithm. In the literature, message-passing GNN (MP-GNN), as the simplest GNN structure, is frequently used as a fast approximation of SB and we find that not all MILPs's SB can be represented with MP-GNN. We precisely define a class of "MP-tractable" MILPs for which MP-GNNs can accurately approximate SB scores. Particularly, we establish a universal approximation theorem: for any data distribution over the MP-tractable class, there always exists an MP-GNN that can approximate the SB score with arbitrarily high accuracy and arbitrarily high probability, which lays a theoretical foundation of the existing works on imitating SB with MP-GNN. For MILPs without the MP-tractability, unfortunately, a similar result is impossible, which can be illustrated by two MILP instances with different SB scores that cannot be distinguished by any MP-GNN, regardless of the number of parameters. Recognizing this, we explore another GNN structure called the second-order folklore GNN (2-FGNN) that overcomes this limitation, and the aforementioned universal approximation theorem can be extended to the entire MILP space using 2-FGNN, regardless of the MP-tractability. A small-scale numerical experiment is conducted to directly validate our theoretical findings.

Shangjin Zhai, Zhichao Ye, Jialin Liu et al.

Recent advances in large reconstruction and generative models have significantly improved scene reconstruction and novel view generation. However, due to compute limitations, each inference with these large models is confined to a small area, making long-range consistent scene generation challenging. To address this, we propose StarGen, a novel framework that employs a pre-trained video diffusion model in an autoregressive manner for long-range scene generation. The generation of each video clip is conditioned on the 3D warping of spatially adjacent images and the temporally overlapping image from previously generated clips, improving spatiotemporal consistency in long-range scene generation with precise pose control. The spatiotemporal condition is compatible with various input conditions, facilitating diverse tasks, including sparse view interpolation, perpetual view generation, and layout-conditioned city generation. Quantitative and qualitative evaluations demonstrate StarGen's superior scalability, fidelity, and pose accuracy compared to state-of-the-art methods. Project page: https://zju3dv.github.io/StarGen.

Yuchen Li, Ziqi Wang, Qingquan Zhang et al.

This survey comprehensively reviews the multi-dimensionality of game scenario diversity, spotlighting the innovative use of procedural content generation and other fields as cornerstones for enriching player experiences through diverse game scenarios. By traversing a wide array of disciplines, from affective modeling and multi-agent systems to psychological studies, our research underscores the importance of diverse game scenarios in gameplay and education. Through a taxonomy of diversity metrics and evaluation methods, we aim to bridge the current gaps in literature and practice, offering insights into effective strategies for measuring and integrating diversity in game scenarios. Our analysis highlights the necessity for a unified taxonomy to aid developers and researchers in crafting more engaging and varied game worlds. This survey not only charts a path for future research in diverse game scenarios but also serves as a handbook for industry practitioners seeking to leverage diversity as a key component of game design and development.

HanQin Cai, Chandra Kundu, Jialin Liu et al.

Robust matrix completion (RMC) is a widely used machine learning tool that simultaneously tackles two critical issues in low-rank data analysis: missing data entries and extreme outliers. This paper proposes a novel scalable and learnable non-convex approach, coined Learned Robust Matrix Completion (LRMC), for large-scale RMC problems. LRMC enjoys low computational complexity with linear convergence. Motivated by the proposed theorem, the free parameters of LRMC can be effectively learned via deep unfolding to achieve optimum performance. Furthermore, this paper proposes a flexible feedforward-recurrent-mixed neural network framework that extends deep unfolding from fix-number iterations to infinite iterations. The superior empirical performance of LRMC is verified with extensive experiments against state-of-the-art on synthetic datasets and real applications, including video background subtraction, ultrasound imaging, face modeling, and cloud removal from satellite imagery.

Jiyuan Pei, Yi Mei, Jialin Liu et al.

Meta-Black-Box Optimization (MetaBBO) garners attention due to its success in automating the configuration and generation of black-box optimizers, significantly reducing the human effort required for optimizer design and discovering optimizers with higher performance than classic human-designed optimizers. However, existing MetaBBO methods conduct one-off training under the assumption that a stationary problem distribution with extensive and representative training problem samples is pre-available. This assumption is often impractical in real-world scenarios, where diverse problems following shifting distribution continually arise. Consequently, there is a pressing need for methods that can continuously learn from new problems encountered on-the-fly and progressively enhance their capabilities. In this work, we explore a novel paradigm of lifelong learning in MetaBBO and introduce LiBOG, a novel approach designed to learn from sequentially encountered problems and generate high-performance optimizers for Black-Box Optimization (BBO). LiBOG consolidates knowledge both across tasks and within tasks to mitigate catastrophic forgetting. Extensive experiments demonstrate LiBOG's effectiveness in learning to generate high-performance optimizers in a lifelong learning manner, addressing catastrophic forgetting while maintaining plasticity to learn new tasks.

Qingquan Zhang, Qiqi Duan, Bo Yuan et al.

Large Language Models (LLMs) have made significant strides in the field of artificial intelligence, showcasing their ability to interact with humans and influence human cognition through information dissemination. However, recent studies have brought to light instances of bias inherent within these LLMs, presenting a critical issue that demands attention. In our research, we delve deeper into the intricate challenge of harmonising accuracy and fairness in the enhancement of LLMs. While improving accuracy can indeed enhance overall LLM performance, it often occurs at the expense of fairness. Overemphasising optimisation of one metric invariably leads to a significant degradation of the other. This underscores the necessity of taking into account multiple considerations during the design and optimisation phases of LLMs. Therefore, we advocate for reformulating the LLM training process as a multi-objective learning task. Our investigation reveals that multi-objective evolutionary learning (MOEL) methodologies offer promising avenues for tackling this challenge. Our MOEL framework enables the simultaneous optimisation of both accuracy and fairness metrics, resulting in a Pareto-optimal set of LLMs. In summary, our study sheds valuable lights on the delicate equilibrium between accuracy and fairness within LLMs, which is increasingly significant for their real-world applications. By harnessing MOEL, we present a promising pathway towards fairer and more efficacious AI technologies.

Qingquan Zhang, Ziqi Wang, Yuchen Li et al.

In recent years, the generation of diverse game levels has gained increasing interest, contributing to a richer and more engaging gaming experience. A number of level diversity metrics have been proposed in literature, which are naturally multi-dimensional, leading to conflicted, complementary, or both relationships among these dimensions. However, existing level generation approaches often fail to comprehensively assess diversity across those dimensions. This paper aims to expand horizons of level diversity by considering multi-dimensional diversity when training generative models. We formulate the model training as a multi-objective learning problem, where each diversity metric is treated as a distinct objective. Furthermore, a multi-objective evolutionary learning framework that optimises multiple diversity metrics simultaneously throughout the model training process is proposed. Our case study on the commonly used benchmark Super Mario Bros. demonstrates that our proposed framework can enhance multi-dimensional diversity and identify a Pareto front of generative models, which provides a range of tradeoffs among playability and two representative diversity metrics, including a content-based one and a player-centered one. Such capability enables decision-makers to make informed choices when selecting generators accommodating a variety of scenarios and the diverse needs of players and designers.

Yuchen Li, Cong Lin, Muhammad Umair Nasir et al.

We introduce GVGAI-LLM, a video game benchmark for evaluating the reasoning and problem-solving capabilities of large language models (LLMs). Built on the General Video Game AI framework, it features a diverse collection of arcade-style games designed to test a model's ability to handle tasks that differ from most existing LLM benchmarks. The benchmark leverages a game description language that enables rapid creation of new games and levels, helping to prevent overfitting over time. Each game scene is represented by a compact set of ASCII characters, allowing for efficient processing by language models. GVGAI-LLM defines interpretable metrics, including the meaningful step ratio, step efficiency, and overall score, to assess model behavior. Through zero-shot evaluations across a broad set of games and levels with diverse challenges and skill depth, we reveal persistent limitations of LLMs in spatial reasoning and basic planning. Current models consistently exhibit spatial and logical errors, motivating structured prompting and spatial grounding techniques. While these interventions lead to partial improvements, the benchmark remains very far from solved. GVGAI-LLM provides a reproducible testbed for advancing research on language model capabilities, with a particular emphasis on agentic behavior and contextual reasoning.

Jialin Liu, Liang Shi, Dingcui Yu

In cost-sensitive deployments, RAID arrays may combine SSDs with different performance levels. Such heterogeneity arises when aging SSDs degrade yet remain usable, or when failed drives are replaced with new devices of explicitly better performance. While this reduces procurement cost, it creates performance challenges: traditional striping mecahnism distributes requests evenly, but slower SSDs become bottlenecks, leaving faster ones underutilized and limiting overall bandwidth to the slowest drive. To address this, we propose HACache (Heterogeneity Adaptive Cache) for read-intensive workloads. HACache introduces high-performance SSDs as read caches to rebalance request distribution. First, we formalize the request diversion problem and solve it formally. Second, to support optimal diversion ratios searching at runtime, HACache adopts a two-phase request diversion ratio adjustment mechanism. Finally, a cache capacity regulation is adopted to adapt quotas for each backend SSD based on hit rates and request diversion needs. This design maximizes bandwidth utilization. Experiments show HACache improves heterogeneous RAID read performance significantly, with bandwidth gains of about 35\% in typical mixed configurations.

Yangshijie Zhang, Xinda Wang, Jialin Liu et al.

With social media growth, users employ stylistic fonts and font-like emoji to express individuality, creating visually appealing text that remains human-readable. However, these fonts introduce hidden vulnerabilities in NLP models: while humans easily read stylistic text, models process these characters as distinct tokens, causing interference. We identify this human-model perception gap and propose a style-based attack, Style Attack Disguise (SAD). We design two sizes: light for query efficiency and strong for superior attack performance. Experiments on sentiment classification and machine translation across traditional models, LLMs, and commercial services demonstrate SAD's strong attack performance. We also show SAD's potential threats to multimodal tasks including text-to-image and text-to-speech generation.

Jialin Liu, Lisang Ding, Stanley Osher et al.

Implicit models, an emerging model class, compute outputs by iterating a single parameter block to a fixed point. This architecture realizes an infinite-depth, weight-tied network that trains with constant memory, significantly reducing memory needs for the same level of performance compared to explicit models. While it is empirically known that these compact models can often match or even exceed larger explicit networks by allocating more test-time compute, the underlying mechanism remains poorly understood. We study this gap through a nonparametric analysis of expressive power. We provide a strict mathematical characterization, showing that a simple and regular implicit operator can, through iteration, progressively express more complex mappings. We prove that for a broad class of implicit models, this process lets the model's expressive power scale with test-time compute, ultimately matching a much richer function class. The theory is validated across three domains: image reconstruction, scientific computing, and operations research, demonstrating that as test-time iterations increase, the complexity of the learned mapping rises, while the solution quality simultaneously improves and stabilizes.

Jiyuan Pei, Yi Mei, Jialin Liu et al.

Recent neural solvers have demonstrated promising performance in learning to solve routing problems. However, existing studies are primarily based on one-off training on one or a set of predefined problem distributions and scales, i.e., tasks. When a new task arises, they typically rely on either zero-shot generalization, which may be poor due to the discrepancies between the new task and the training task(s), or fine-tuning the pretrained solver on the new task, which possibly leads to catastrophic forgetting of knowledge acquired from previous tasks. This paper explores a novel lifelong learning paradigm for neural VRP solvers, where multiple tasks with diverse distributions and scales arise sequentially over time. Solvers are required to effectively and efficiently learn to solve new tasks while maintaining their performance on previously learned tasks. Consequently, a novel framework called Lifelong Learning Router with Behavior Consolidation (LLR-BC) is proposed. LLR-BC consolidates prior knowledge effectively by aligning behaviors of the solver trained on a new task with the buffered ones in a decision-seeking way. To encourage more focus on crucial experiences, LLR-BC assigns greater consolidated weights to decisions with lower confidence. Extensive experiments on capacitated vehicle routing problems and traveling salesman problems demonstrate LLR-BC's effectiveness in training high-performance neural solvers in a lifelong learning setting, addressing the catastrophic forgetting issue, maintaining their plasticity, and improving zero-shot generalization ability.

Qingquan Zhang, Yuchen Li, Bo Yuan et al.

Large language models (LLMs) have demonstrated tremendous potential in game playing, while little attention has been paid to their ethical implications in those contexts. This work investigates and analyses the ethical considerations of applying LLMs in game playing, using Werewolf, also known as Mafia, as a case study. Gender bias, which affects game fairness and player experience, has been observed from the behaviour of LLMs. Some roles, such as the Guard and Werewolf, are more sensitive than others to gender information, presented as a higher degree of behavioural change. We further examine scenarios in which gender information is implicitly conveyed through names, revealing that LLMs still exhibit discriminatory tendencies even in the absence of explicit gender labels. This research showcases the importance of developing fair and ethical LLMs. Beyond our research findings, we discuss the challenges and opportunities that lie ahead in this field, emphasising the need for diving deeper into the ethical implications of LLMs in gaming and other interactive domains.

Chengpeng Hu, Ziming Wang, Bo Yuan et al.

Dynamic material handling (DMH) involves the assignment of dynamically arriving material transporting tasks to suitable vehicles in real time for minimising makespan and tardiness. In real-world scenarios, historical task records are usually available, which enables the training of a decision policy on multiple instances consisting of historical records. Recently, reinforcement learning has been applied to solve DMH. Due to the occurrence of dynamic events such as new tasks, adaptability is highly required. Solving DMH is challenging since constraints including task delay should be satisfied. A feedback is received only when all tasks are served, which leads to sparse reward. Besides, making the best use of limited computational resources and historical records for training a robust policy is crucial. The time allocated to different problem instances would highly impact the learning process. To tackle those challenges, this paper proposes a novel adaptive constrained evolutionary reinforcement learning (ACERL) approach, which maintains a population of actors for diverse exploration. ACERL accesses each actor for tackling sparse rewards and constraint violation to restrict the behaviour of the policy. Moreover, ACERL adaptively selects the most beneficial training instances for improving the policy. Extensive experiments on eight training and eight unseen test instances demonstrate the outstanding performance of ACERL compared with several state-of-the-art algorithms. Policies trained by ACERL can schedule the vehicles while fully satisfying the constraints. Additional experiments on 40 unseen noised instances show the robust performance of ACERL. Cross-validation further presents the overall effectiveness of ACREL. Besides, a rigorous ablation study highlights the coordination and benefits of each ingredient of ACERL.

Shiying Hu, Zengrong Huang, Chengpeng Hu et al.

Recently, procedural content generation has exhibited considerable advancements in the domain of 2D game level generation such as Super Mario Bros. and Sokoban through large language models (LLMs). To further validate the capabilities of LLMs, this paper explores how LLMs contribute to the generation of 3D buildings in a sandbox game, Minecraft. We propose a Text to Building in Minecraft (T2BM) model, which involves refining prompts, decoding interlayer representation and repairing. Facade, indoor scene and functional blocks like doors are supported in the generation. Experiments are conducted to evaluate the completeness and satisfaction of buildings generated via LLMs. It shows that LLMs hold significant potential for 3D building generation. Given appropriate prompts, LLMs can generate correct buildings in Minecraft with complete structures and incorporate specific building blocks such as windows and beds, meeting the specified requirements of human users.

Ziang Chen, Xiaohan Chen, Jialin Liu et al.

Quadratic programming (QP) is the most widely applied category of problems in nonlinear programming. Many applications require real-time/fast solutions, though not necessarily with high precision. Existing methods either involve matrix decomposition or use the preconditioned conjugate gradient method. For relatively large instances, these methods cannot achieve the real-time requirement unless there is an effective preconditioner. Recently, graph neural networks (GNNs) opened new possibilities for QP. Some promising empirical studies of applying GNNs for QP tasks show that GNNs can capture key characteristics of an optimization instance and provide adaptive guidance accordingly to crucial configurations during the solving process, or directly provide an approximate solution. However, the theoretical understanding of GNNs in this context remains limited. Specifically, it is unclear what GNNs can and cannot achieve for QP tasks in theory. This work addresses this gap in the context of linearly constrained QP tasks. In the continuous setting, we prove that message-passing GNNs can universally represent fundamental properties of convex quadratic programs, including feasibility, optimal objective values, and optimal solutions. In the more challenging mixed-integer setting, while GNNs are not universal approximators, we identify a subclass of QP problems that GNNs can reliably represent.

Chengpeng Hu, Jialin Liu, Xin Yao

Recently, evolutionary reinforcement learning has obtained much attention in various domains. Maintaining a population of actors, evolutionary reinforcement learning utilises the collected experiences to improve the behaviour policy through efficient exploration. However, the poor scalability of genetic operators limits the efficiency of optimising high-dimensional neural networks.To address this issue, this paper proposes a novel cooperative coevolutionary reinforcement learning (CoERL) algorithm. Inspired by cooperative coevolution, CoERL periodically and adaptively decomposes the policy optimisation problem into multiple subproblems and evolves a population of neural networks for each of the subproblems. Instead of using genetic operators, CoERL directly searches for partial gradients to update the policy. Updating policy with partial gradients maintains consistency between the behaviour spaces of parents and offspring across generations.The experiences collected by the population are then used to improve the entire policy, which enhances the sampling efficiency.Experiments on six benchmark locomotion tasks demonstrate that CoERL outperforms seven state-of-the-art algorithms and baselines.Ablation study verifies the unique contribution of CoERL's core ingredients.