arXiv subjects

Tianyuan Liu

Publications and source records attributed to Tianyuan Liu.

At least 19 recordsLinked to original sources

arXiv2026

Execution-Aware Pre-Execution Ranking for Grasp-Conditioned Robotic Placement

A geometrically valid placement can still be difficult to execute because the selected grasp changes the required end-effector pose, collision geometry, and transport motion. Placement is formulated as a pre-execution ranking problem in which supplied grasp-placement candidates are scored before planning. The model combines a typed target-conditioned point cloud with three pose descriptors and hierarchical heads for planning success and execution success conditioned on planning. On a 30-object, 1,235-scene dataset with scene-group-held-out splits, three-seed top-1 success on covered test groups reaches 85.63 +/- 1.08% for joint selection and 79.84 +/- 0.16% for fixed-target ranking. For the designated frozen seed-42 checkpoint, top-1 success improves from 72.84% to 85.78% over full-pool cuMotion for joint ranking and from 59.65% to 79.67% for fixed-target ranking. Frozen transfer to xArm7/MoveIt requires no xArm-specific retraining. Across 27 locked cases, 13 complete end to end (48.15%). Of the 16 cases that pass Top-5 preflight and begin execution, 13 succeed (81.25%). Candidate-level deployment-feasibility prediction reaches 81.25% recall, 85.20% specificity, and 83.23% balanced accuracy.

cs.RO↗

arXiv2026

Hybrid Residual Reinforcement Learning for Contact-Rich Robotic Book Insertion

Placing a grasped book into a tight shelf is a compact but difficult contact-rich control problem: millimetre-scale pose error can turn a geometrically valid approach into jamming, failed release, or incomplete seating. We study this final phase after grasp acquisition and global approach, and ask how control authority should be divided between known geometry and learned behaviour. Our method retains a nominal task-space controller for structured insertion and seating, while residual PPO supplies bounded local corrections and decides when to release. Only the brief open-retreat-reclose transition is scripted. For the final policy used on hardware, a deployment-matched simulation evaluation over 512 fixed conditions yields 98.50 percent mean success (0.23 percentage-point sample SD) across three independent training runs, compared with 37.89 percent for nominal control. On the physical xArm7, 60 trials over 30 matched conditions show the same qualitative advantage: residual control raises success from 26.7 percent to 63.3 percent, reduces failures from 22 to 11, and wins 13 of the 15 matched conditions in which the two controllers differ. Robustness tests show that performance remains above 87 percent under initialization perturbations up to 1.5x, while very tight clearances expose the geometric limit of local correction. These results support a hybrid design in which geometry preserves reliable task structure and learning is concentrated on the contact-sensitive behaviour that fixed rules handle poorly.

cs.RO↗

arXiv2026

A panoramic aerodynamic performance prediction method for turbomachinery cascades using transformer-enhanced neural operator

To enable flexible and rapid aerodynamic performance evaluation in turbomachinery design, this paper proposes a panoramic performance prediction framework. Unlike most previous prediction models that directly predict the objective functions of interest, our approach first predicts the basic parameters of the Navier-Stokes equations, such as temperature, pressure, and density. Utilizing these basic physical quantities, it subsequently predicts key performance parameters of the turbine stage meridian plane. By adopting this methodology, our proposed panoramic performance prediction framework functions similarly to a CFD simulator, capable of predicting various objective of interest to the designers. To enhance prediction accuracy, a transformer-enhanced neural operator (TNO) is introduced within this framework. Using the Rotor 37 blades as a reference, the proposed TNO is trained to predict the performance of a transonic compressor blade in the meridian plane. The TNO can accurately predict total quantities such as isentropic efficiency, mass flow, and distributions of total pressure ratio. Remarkably, the prediction error of TNO is observed to be smaller than that of state-of-the-art deep learning operators such as the FNO and DeepONet. Furthermore, the TNO is applied to downstream tasks, including sensitivity analysis and optimization of various objective functions. The results confirm that the TNO can operate almost like a CFD simulator, while reducing the computational cost of downstream tasks by four orders of magnitude. The effectiveness and reliability of the proposed TNO for solving different kinds of downstream tasks have been well demonstrated.

cs.LG↗

arXiv2026

AI Surrogate Modeling for Real-Time Tokamak Equilibrium Prediction: Benchmarking Neural Architectures and Validation on EXL-50U

Fast and reliable plasma equilibrium prediction is essential for real-time tokamak operation and control, but conventional Grad-Shafranov (GS) solvers are often too costly for real-time deployment. We develop an AI surrogate framework and benchmark five architectures (MLP, CNN, FNO, Transformer, and KAN) on a numerical GS database with 100,000 IID and 10,000 OOD samples. Under a unified protocol, we evaluate accuracy, inference efficiency, model scaling, and robustness. We also establish device-level validation on the EXL-50U tokamak by linking numerical GS solutions, surrogate predictions, and the standard Shape Editor reference to assess simulation-to-device consistency. The surrogates achieve errors of $10^{-3}$-$10^{-2}$ relative to GS solutions, while the GS-to-device discrepancy remains at $10^{-3}$. Transformer gives the best IID accuracy, whereas CNN offers the best balance of accuracy, robustness, and speed, reaching 0.7 ms TensorRT latency. On unseen plasma geometries and parameter regimes, CNN and FNO show the strongest extrapolation stability, with 4%-5% relative $L_2$ error, while models with weaker inductive biases degrade more substantially. Scaling data and model capacity improves interpolation but not necessarily extrapolation, revealing a trade-off between capacity and OOD generalization. Overall, this work provides a systematic, device-consistent benchmark for AI-based GS prediction and practical guidance for selecting reliable surrogates for real-time plasma control and fusion applications.

physics.plasm-ph↗

arXiv2026

State-Space Model-Enabled Reinforcement Learning for Magnetic Configuration Controlon EXL-50U

Accurate feedback control of the plasma current ($I_p$) and centroid position $(R_c,Z_c)$ is essential for the stable operation of spherical torus (ST) plasmas. Conventional proportional-integral-derivative (PID) controllers require extensive manual tuning and struggle with the fast, strongly coupled dynamics that arise as plasma performance improves. Reinforcement learning (RL) has recently emerged as a promising alternative to such complex magnetic control problems, yet its practical deployment on ST devices remains challenging. This paper presents a practical RL controller for the EXL-50U ST, trained within a rigid RZIP state-space model (SSM) that enables efficient offline policy learning. A lightweight plasma position reconstructor is developed to estimate $(R_c,Z_c)$ from magnetic probe signals within the real-time control cycle. The trained policy is seamlessly deployed on the EXL-50U plasma control system, achieving stable regulation of $I_p$ and $(R_c,Z_c)$ and sustaining discharges up to 650 ms under RL control. These results demonstrate the feasibility and practical potential of model-informed RL for magnetic configuration control in ST devices, offering a promising direction beyond conventional PID-based schemes.

physics.plasm-ph↗

arXiv2026

Power-law-anchored residual learning for H-mode energy confinement time in tokamaks: interpolation and parameter-defined extrapolation

Reliable prediction of the energy confinement time is essential for magnetic-confinement fusion. Conventional power-law scalings provide constrained extrapolation trends but cannot represent complex nonlinearities, whereas neural networks interpolate accurately but may behave unpredictably outside the training distribution. We propose a unified power-law-anchored residual-learning framework in which a frozen empirical power-law scaling supplies the global trend and a nonlinear model learns only the systematic residual in logarithmic space. PLR-KAN is developed as the primary implementation, while a parameter-matched PLR-MLP serves as a controlled architecture replacement. Using the ITPA DB5.2.3 H-mode confinement database, we evaluate interpolation and parameter-defined held-out cohorts over ten complete training pipelines. PLR-KAN retains near-best interpolation accuracy, achieving R2=0.9671+/-0.0027, while substantially improving the stability of direct KAN under parameter-defined distribution shifts. It outperforms direct KAN across all five non-epsilon single-parameter-defined cohorts and the core-five joint cohort, reaching R2=0.9263+/-0.0157 in the latter. Results from PLR-MLP further demonstrate that the benefit of power-law anchoring is not specific to KAN, although the effectiveness of residual transfer remains architecture and direction dependent. As an exploratory extension, a Mahalanobis-distance-based prediction-time gate improves stability in selected shifted regions but is not universally beneficial and cannot compensate for missing device or physics-regime coverage. Overall, power-law-anchored residual learning provides a practical balance between nonlinear interpolation capability and empirically constrained extrapolation behavior.

physics.plasm-ph↗

arXiv2026

Advantage-level Aggregation Reinforcement Learning for X-point Target Magnetic Configuration Control in an EXL-50U Experiment-Calibrated Simulation Environment

Managing divertor heat loads is a central challenge for compact, high-power tokamaks. To increase local flux expansion and decouple the dissipation volume from the core, EHL-2 adopts the X-point target (XPT) divertor. This requires the secondary X-point to remain on the divertor leg; displacement degrades the topology and exhaust geometry. Current experiments, including EXL-50U discharges, rely on precomputed feedforward waveforms with PID loops on global quantities. Lacking dedicated closed-loop feedback for the secondary null, XPT operation is repeatable but not routine. We formulate XPT feedback as a multi-objective reinforcement learning (RL) control problem in a free-boundary environment calibrated to EXL-50U discharge #13906. To address strong coupling among plasma current, shape, and null constraints - where reward scalarisation collapses objective-specific temporal credit - we develop Advantage Aggregation (AdvA). AdvA preserves objective-wise temporal credit before worst-objective-aware nonlinear scalarisation and introduces a residual correction to policy updates. AdvA-PPO is evaluated against Reward-PPO and a feedforward-plus-PID baseline under nominal operation, measurement uncertainties, and unseen initial equilibria. On a 500 ms rollout, AdvA-PPO raises the mean worst-channel score from 0.23 to 0.81 over Reward-PPO, reducing X-point flux RMSE by ~20x. Under combined measurement uncertainties, it is the only learned controller completing the horizon while retaining a usable XPT shape. Multi-initialization fine-tuning enables a single AdvA-PPO policy to complete full-horizon operation across divertor and limiter initial equilibria. These results provide a simulation-based foundation for future real-time XPT validation on EXL-50U.

physics.plasm-ph↗

arXiv2026

Reinforcement learning for vertical position control on the EXL-50U spherical tokamak

Vertical position control is essential for sustaining high-performance operation in spherical tokamaks, where increased plasma elongation introduces stringent requirements on fast and robust stabilization. This work presents an experimentally validated reinforcement-learning(RL)-based vertical position control framework for the EXL-50U spherical tokamak. A high-fidelity discharge-reconstructed simulation environment is developed by integrating physics-based plasma-circuit models with experimental equilibrium information, enabling systematic controller synthesis and sim-to-real evaluation. Within this framework, RL is benchmarked in simulation against operational proportional--integral--derivative (PID) and model-based linear quadratic regulator (LQR) controllers under identical plant dynamics, actuator constraints, and measurement imperfections.Simulation results show that RL achieves tracking accuracy comparable to PID with consistently lower vertical-stabilization coil effort, while lightweight integral compensation improves robustness against residual model--plant mismatch. The RL controller is subsequently deployed on EXL-50U for closed-loop experiments. Across more than ten discharges with RL takeover, stable vertical regulation is achieved within the controlled windows. For seven representative discharges, RL maintains millimetre-scale tracking accuracy comparable to the operational PID controller (MAE typically ~ 1-5 mm) while consistently reducing actuator effort. These results demonstrate the feasibility of learning-based plasma control on a real spherical tokamak and establish a practical pathway toward future fusion control systems.

physics.plasm-ph↗

arXiv2026

Interpretable and Sparse Linear Attention with Decoupled Membership-Subspace Modeling via MCR2 Objective

Maximal Coding Rate Reduction (MCR2)-driven white-box transformer, grounded in structured representation learning, unifies interpretability and efficiency, providing a reliable white-box solution for visual modeling. However, in existing designs, tight coupling between "membership matrix" and "subspace matrix U" in MCR2 causes redundant coding under incorrect token projection. To this end, we decouple the functional relationship between the "membership matrix" and "subspaces U" in the MCR2 objective and derive an interpretable sparse linear attention operator from unrolled gradient descent of the optimized objective. Specifically, we propose to directly learn the membership matrix from inputs and subsequently derive sparse subspaces from the fullspace S. Consequently, gradient unrolling of the optimized MCR2 objective yields an interpretable sparse linear attention operator: Decoupled Membership-Subspace Attention (DMSA). Experimental results on visual tasks show that simply replacing the attention module in Token Statistics Transformer (ToST) with DMSA (we refer to as DMST) not only achieves a faster coding reduction rate but also outperforms ToST by 1.08%-1.45% in top-1 accuracy on the ImageNet-1K dataset. Compared with vanilla Transformer architectures, DMST exhibits significantly higher computational efficiency and interpretability.

cs.CV↗

arXiv2025

Optimizing Robotic Placement via Grasp-Dependent Feasibility Prediction

In this paper, we study whether inexpensive, physics-free supervision can reliably prioritize grasp-place candidates for budget-aware pick-and-place. From an object's initial pose, target pose, and a candidate grasp, we generate two path-aware geometric labels: path-wise inverse kinematics (IK) feasibility across a fixed approach-grasp-lift waypoint template, and a transit collision flag from mesh sweeps along the same template. A compact dual-output MLP learns these signals from pose encodings, and at test time its scores rank precomputed candidates for a rank-then-plan policy under the same IK gate and planner as the baseline. Although learned from cheap labels only, the scores transfer to physics-enabled executed trajectories: at a fixed planning budget the policy finds successful paths sooner with fewer planner calls while keeping final success on par or better. This work targets a single rigid cuboid with side-face grasps and a fixed waypoint template, and we outline extensions to varied objects and richer waypoint schemes.

cs.RO↗

arXiv2025

Physics-informed Neural Operator Learning for Nonlinear Grad-Shafranov Equation

As artificial intelligence emerges as a transformative enabler for fusion energy commercialization, fast and accurate solvers become increasingly critical. In magnetic confinement nuclear fusion, rapid and accurate solution of the Grad-Shafranov equation (GSE) is essential for real-time plasma control and analysis. Traditional numerical solvers achieve high precision but are computationally prohibitive, while data-driven surrogates infer quickly but fail to enforce physical laws and generalize poorly beyond training distributions. To address this challenge, we present a Physics-Informed Neural Operator (PINO) that directly learns the GSE solution operator, mapping shape parameters of last closed flux surface to equilibrium solutions for realistic nonlinear current profiles. Comprehensive benchmarking of five neural architectures identifies the novel Transformer-KAN (Kolmogorov-Arnold Network) Neural Operator (TKNO) as achieving highest accuracy (0.25% mean L2 relative error) under supervised training (only data-driven). However, all data-driven models exhibit large physics residuals, indicating poor physical consistency. Our unsupervised training can reduce the residuals by nearly four orders of magnitude through embedding physics-based loss terms without labeled data. Critically, semi-supervised learning--integrating sparse labeled data (100 interior points) with physics constraints--achieves optimal balance: 0.48% interpolation error and the most robust extrapolation performance (4.76% error, 8.9x degradation factor vs 39.8x for supervised models). Accelerated by TensorRT optimization, our models enable millisecond-level inference, establishing PINO as a promising pathway for next-generation fusion control systems.

physics.plasm-ph↗

arXiv2025

TADT-CSA: Temporal Advantage Decision Transformer with Contrastive State Abstraction for Generative Recommendation

With the rapid advancement of Transformer-based Large Language Models (LLMs), generative recommendation has shown great potential in enhancing both the accuracy and semantic understanding of modern recommender systems. Compared to LLMs, the Decision Transformer (DT) is a lightweight generative model applied to sequential recommendation tasks. However, DT faces challenges in trajectory stitching, often producing suboptimal trajectories. Moreover, due to the high dimensionality of user states and the vast state space inherent in recommendation scenarios, DT can incur significant computational costs and struggle to learn effective state representations. To overcome these issues, we propose a novel Temporal Advantage Decision Transformer with Contrastive State Abstraction (TADT-CSA) model. Specifically, we combine the conventional Return-To-Go (RTG) signal with a novel temporal advantage (TA) signal that encourages the model to capture both long-term returns and their sequential trend. Furthermore, we integrate a contrastive state abstraction module into the DT framework to learn more effective and expressive state representations. Within this module, we introduce a TA-conditioned State Vector Quantization (TAC-SVQ) strategy, where the TA score guides the state codebooks to incorporate contextual token information. Additionally, a reward prediction network and a contrastive transition prediction (CTP) network are employed to ensure the state codebook preserves both the reward information of the current state and the transition information between adjacent states. Empirical results on both public datasets and an online recommendation system demonstrate the effectiveness of the TADT-CSA model and its superiority over baseline methods.

cs.IR↗

arXiv2024

Prediction of turbulent channel flow using Fourier neural operator-based machine-learning strategy

Fast and accurate predictions of turbulent flows are of great importance in the science and engineering field. In this paper, we investigate the implicit U-Net enhanced Fourier neural operator (IUFNO) in the stable prediction of long-time dynamics of three-dimensional (3D) turbulent channel flows. The trained IUFNO models are tested in the large-eddy simulations (LES) at coarse grids for three friction Reynolds numbers: $Re_τ\approx180$, $395$ and $590$. The adopted near-wall mesh grids are tangibly coarser than the general requirements for wall-resolved LES. Compared to the original Fourier neural operator (FNO), the implicit FNO (IFNO) and U-Net enhanced FNO (UFNO), the IUFNO model has a much better long-term predictive ability. The numerical experiments show that the IUFNO framework outperforms the traditional dynamic Smagorinsky model (DSM) and the wall-adapted local eddy-viscosity (WALE) model in the predictions of a variety of flow statistics and structures, including the mean and fluctuating velocities, the probability density functions (PDFs) and joint PDF of velocity fluctuations, the Reynolds stress profile, the kinetic energy spectrum, and the Q-criterion (vortex structures). Meanwhile, the trained IUFNO models are computationally much faster than the traditional LES models. Thus, the IUFNO model is a promising approach for the fast prediction of wall-bounded turbulent flow.

physics.flu-dyn↗

arXiv2024

Fourier neural operator for large eddy simulation of compressible Rayleigh-Taylor turbulence

The Fourier neural operator (FNO) framework is applied to the large eddy simulation (LES) of three-dimensional compressible Rayleigh-Taylor (RT) turbulence with miscible fluids at Atwood number $A_t=0.5$, stratification parameter $Sr=1.0$, and Reynolds numbers $Re=10000$ and 30000. The FNO model is first used for predicting three-dimensional compressible turbulence. The different magnitudes of physical fields are normalized using root mean square values for an easier training of FNO models. In the \emph{a posteriori} tests, the FNO model outperforms the velocity gradient model (VGM), the dynamic Smagorinsky model (DSM), and implicit large eddy simulation (ILES) in predicting various statistical quantities and instantaneous structures, and is particularly superior to traditional LES methods in predicting temperature fields and velocity divergence. Moreover, the computational efficiency of the FNO model is much higher than that of traditional LES methods. FNO models trained with short-time, low Reynolds number data exhibit a good generalization performance on longer-time predictions and higher Reynolds numbers in the \emph{a posteriori} tests.

physics.flu-dyn↗

arXiv2024

$\text{Alpha}^2$: Discovering Logical Formulaic Alphas using Deep Reinforcement Learning

Alphas are pivotal in providing signals for quantitative trading. The industry highly values the discovery of formulaic alphas for their interpretability and ease of analysis, compared with the expressive yet overfitting-prone black-box alphas. In this work, we focus on discovering formulaic alphas. Prior studies on automatically generating a collection of formulaic alphas were mostly based on genetic programming (GP), which is known to suffer from the problems of being sensitive to the initial population, converting to local optima, and slow computation speed. Recent efforts employing deep reinforcement learning (DRL) for alpha discovery have not fully addressed key practical considerations such as alpha correlations and validity, which are crucial for their effectiveness. In this work, we propose a novel framework for alpha discovery using DRL by formulating the alpha discovery process as program construction. Our agent, $\text{Alpha}^2$, assembles an alpha program optimized for an evaluation metric. A search algorithm guided by DRL navigates through the search space based on value estimates for potential alpha outcomes. The evaluation metric encourages both the performance and the diversity of alphas for a better final trading strategy. Our formulation of searching alphas also brings the advantage of pre-calculation dimensional analysis, ensuring the logical soundness of alphas, and pruning the vast search space to a large extent. Empirical experiments on real-world stock markets demonstrates $\text{Alpha}^2$'s capability to identify a diverse set of logical and effective alphas, which significantly improves the performance of the final trading strategy. The code of our method is available at https://github.com/x35f/alpha2.

q-fin.CP↗

arXiv2024

A transformer-based neural operator for large-eddy simulation of turbulence

Predicting the large-scale dynamics of three-dimensional (3D) turbulence is challenging for machine learning approaches. This paper introduces a transformer-based neural operator (TNO) to achieve precise and efficient predictions in the large-eddy simulation (LES) of 3D turbulence. The performance of the proposed TNO model is systematically tested and compared with LES using classical sub-grid scale (SGS) models, including the dynamic Smagorinsky model (DSM) and the dynamic mixed model (DMM), as well as the original Fourier neural operator (FNO) model, in homogeneous isotropic turbulence (HIT) and free-shear turbulent mixing layer. The numerical simulations comprehensively evaluate the performance of these models on a variety of flow statistics, including the velocity spectrum, the probability density functions (PDFs) of vorticity, the PDFs of velocity increments, the evolution of turbulent kinetic energy, and the iso-surface of the Q-criterion. The results indicate that the accuracy of the TNO model is comparable to the LES with DSM model, and outperforms the FNO model and LES using DMM in HIT. In the free-shear turbulence, the TNO model exhibits superior accuracy compared to other models. Moreover, the TNO model has fewer parameters than the FNO model and enables long-term stable predictions, which the FNO model cannot achieve. The well-trained TNO model is significantly faster than traditional LES with DSM and DMM models, and can be generalized to higher Taylor-Reynolds number cases, indicating its strong potential for 3D nonlinear engineering applications.

physics.flu-dyn↗

arXiv2023

Rotating Objects via In-Hand Pivoting using Vision, Force and Touch

We propose a robotic manipulation system that can pivot objects on a surface using vision, wrist force and tactile sensing. We aim to control the rotation of an object around the grip point of a parallel gripper by allowing rotational slip, while maintaining a desired wrist force profile. Our approach runs an end-effector position controller and a gripper width controller concurrently in a closed loop. The position controller maintains a desired force using vision and wrist force. The gripper controller uses tactile sensing to keep the grip firm enough to prevent translational slip, but loose enough to induce rotational slip. Our sensor-based control approach relies on matching a desired force profile derived from object dimensions and weight and vision-based monitoring of the object pose. The gripper controller uses tactile sensors to detect and prevent translational slip by tightening the grip when needed. Experimental results where the robot was tasked with rotating cuboid objects 90 degrees show that the multi-modal pivoting approach was able to rotate the objects without causing lift or slip, and was more energy-efficient compared to using a single sensor modality and to pick-and-place. While our work demonstrated the benefit of multi-modal sensing for the pivoting task, further work is needed to generalize our approach to any given object.

cs.RO↗

arXiv2023

Divide and Conquer in Video Anomaly Detection: A Comprehensive Review and New Approach

Video anomaly detection is a complex task, and the principle of "divide and conquer" is often regarded as an effective approach to tackling intricate issues. It's noteworthy that recent methods in video anomaly detection have revealed the application of the divide and conquer philosophy (albeit with distinct perspectives from traditional usage), yielding impressive outcomes. This paper systematically reviews these literatures from six dimensions, aiming to enhance the use of the divide and conquer strategy in video anomaly detection. Furthermore, based on the insights gained from this review, a novel approach is presented, which integrates human skeletal frameworks with video data analysis techniques. This method achieves state-of-the-art performance on the ShanghaiTech dataset, surpassing all existing advanced methods.

cs.CV↗