arXiv subjects

Tianyi Ding

Publications and source records attributed to Tianyi Ding.

12 recordsLinked to original sources

arXiv2026

Training-Free Motion-Corrected 3D Brain MRI Reconstruction via Event-Guided References

Head motion causes artefacts in 3D brain MRI and can bias structural measurements. Model-based correction alternates between image and motion estimation, but artefacts in the reconstructed reference can reinforce motion errors. Learned priors can reduce these artefacts but require pretraining and may add test-time computation. We propose EventMC (event-guided motion correction), a training-free method for interleaved Cartesian acquisitions. It iteratively detects motion events and estimates motion using reference images reconstructed from stable intervals. It then refines the image and motion from low to high spatial frequencies. In a fair comparison in simulation (same reconstruction), EventMC's motion estimates improve mean PSNR by up to 4.0 dB over those of the learning-based MotionTTT. A single run of EventMC is up to 7.5 times faster than MotionTTT with less than half the GPU memory. On in vivo motion-corrupted scans, EventMC achieves similar mean image quality to MotionTTT without any training. Code is available at https://github.com/bayesbreeze/EventMC.

cs.CV↗

arXiv2026

A Hybrid SMT-NRA Solver: Integrating 2D Cell-Jump-Based Local Search, MCSAT and OpenCAD

In this paper, we propose a hybrid framework for Satisfiability Modulo the Theory of Nonlinear Real Arithmetic (SMT-NRA for short). First, we introduce a two-dimensional cell-jump move, called \emph{2D-cell-jump}, generalizing the key operation, cell-jump, of the local search method for SMT-NRA. In contrast to the existing cell-jump, which updates assignments exclusively along a given line, the 2D-cell-jump enables search over an arbitrarily selected plane. Then, we propose an extended local search framework, named \emph{2D-LS} (following the existing local search framework, LS, for SMT-NRA). Finally, we present a hybrid framework for SMT-NRA integrating 2D-LS, MCSAT and OpenCAD, to improve search efficiency through information exchange. The experimental results demonstrate improvements in local search performance, highlighting the effectiveness of the proposed methods.

cs.AI↗

arXiv2026

DeepPV-MRF: End-to-End Partial-Volume Fraction Estimation from MR Fingerprinting Time Series

Background: Estimating subvoxel tissue fractions from magnetic resonance fingerprinting (MRF) is challenging. Conventional dictionary matching and segmentation pipelines can propagate errors from intermediate parametric maps into tissue fraction estimates. Purpose: We introduce DeepPV-MRF, a deep learning framework that directly estimates voxel-wise tissue fractions from raw MRF signals. Methods: A cascaded spatial-temporal network was trained on subject-derived Bloch simulations with radial k-space undersampling. The network combines temporal encoding and spatial U-Net refinement to estimate cerebrospinal fluid (CSF), gray matter (GM), and white matter (WM) fractions without intermediate T1/T2 maps. We compared DeepPV-MRF with partial-volume dictionary matching (PV-DM) and pseudoinverse estimation (PV-Pinv) using simulated and in vivo data. Results: DeepPV-MRF outperformed both methods across acquisition settings. In single-shot simulations, CSF/WM/GM mean absolute errors were 0.0418/0.0744/0.0757, versus 0.1281/0.2443/0.2958 for PV-DM and 0.2021/0.2793/0.2630 for PV-Pinv. Corresponding structural similarity values were 0.960/0.916/0.922. In vivo results showed sharper anatomical boundaries and smoother tissue transitions, while cross-shot analysis indicated improved consistency across volunteers. Conclusions: DeepPV-MRF enables accurate and efficient direct tissue fraction estimation from MRF signals, particularly with limited shots. Further validation is needed before clinical deployment.

eess.SP↗

arXiv2026

PEARS: Physical-Prior-Guided Efficient Adaptation via Failure Reasoning and Diffusion Steering for Tactile Manipulation

Pretrained robotic policies can suffer substantial performance degradation under out-of-distribution (OOD) conditions encountered during deployment, motivating post-training through real-world interaction. However, reinforcement-learning (RL)-based post-training typically requires substantial environment interactions, a burden that is especially significant in manipulation, where each trial can be slow, costly, or destructive. Therefore, we present PEARS, a physics-prior-guided hybrid RL framework for sample-efficient online adaptation of pretrained policies with tactile feedback. After each episode, its physics-guided force reasoning (PFR) module uses physical priors encoded in a vision-language model (VLM) to diagnose failures from the visual outcome and tactile interaction history and update task-appropriate contact-force bounds. A high-frequency hybrid force-position controller then enforces these bounds during contact. Complementarily, tactile-conditioned diffusion steering reinforcement learning adjusts the latent noise of the frozen flow-matching policy to correct errors in free-space motion and contact timing without updating the base model. In simulation, PEARS improves success rates by 12.4-37.4 percentage points over the strongest per-task baselines. PEARS also reduces the number of interaction episodes required for a certain success threshold by up to 53.2% relative to the fastest baseline. In real-world experiments, PEARS achieves success rates of 95% on Whiteboard Erasing and 90% on Pipette Liquid Aspiration. These results show that combining the PFR module with policy steering can accelerate adaptation while reducing costly interactions. The project website is available at https://song-kun.github.io/pears.

cs.RO↗

arXiv2026

MeSD: Multi-Evidence Self-Distillation for VideoLLM

While reinforcement learning with verifiable rewards provides reliable outcome supervision for VideoLLMs, sequence-level rewards offer limited token-level guidance. On-policy self-distillation addresses this limitation by conditioning a self-teacher on privileged information to provide dense token-level supervision. However, aggregating heterogeneous evidence within a single teacher context obscures cross-evidence agreement and conflict. A further challenge lies in determining whether teacher guidance should refine reward-based updates or provide corrective supervision for failed trajectories. To address these issues, we propose MeSD, a multi-evidence self-distillation framework for VideoLLMs. MeSD constructs three evidence-conditioned teachers with shared parameters, using the ground-truth answer as a common semantic context while separately incorporating temporal and spatial evidence. Given the same student-generated prefixes, MeSD evaluates evidence-specific preferences relative to the Answer Teacher and fuses teacher-common preferences with gated teacher-specific residuals. Furthermore, MeSD introduces Verification-Guided Optimization to classify trajectories as Success, Failure, or Indeterminate. For Success and Indeterminate trajectories, MeSD refines token-level advantage magnitudes while preserving reward-derived signs. For verified failure trajectories that contain the required evidence, MeSD applies failure-conditioned distillation, using reverse-KL correction toward the fused distribution. Experiments on multiple video benchmarks demonstrate consistent gains over reinforcement learning and self-distillation baselines.

cs.CV↗

arXiv2026

Probing Neutrino Flavor Composition with the Glashow Resonance at Tau Air-Shower Neutrino Telescopes

The flavor composition of high-energy astrophysical neutrinos encodes information about their production and propagation. The Glashow resonance, $\barν_e + e^-\to W^-$, provides a unique way to distinguish antineutrinos from neutrinos and thereby extends the reach of flavor composition studies. Proposed tau air-shower neutrino telescopes target Earth-skimming and mountain-skimming $ν_τ$ above a PeV, but through the decay $W^-\to \barν_τ+τ^-$ they are also sensitive to $\barν_e$. These experiments can therefore measure the ratio of $\barν_e$ to $ν_τ+\barν_τ$ fluxes. We evaluate this prospect with explicit simulations and project sensitivities for TAMBO and TRINITY, assuming 10 years of operation. We find that mountain-skimming geometries yield substantially higher $\barν_e$ acceptance than Earth-skimming ones due to the shorter path length in rock. For standard astrophysical source scenarios, our projections show that differentiating $pp$ and $pγ$ production, including their muon-damped scenarios, is challenging with a standalone measurement by tau air-shower experiments in their currently designed configurations, though optimistically the flux ratio can be constrained to $\lesssim$2 at 1$σ$. A $\barν_e$-rich flux, as expected from neutron-decay sources or from certain new physics models, would stand out from the standard pion-production scenarios and can otherwise be constrained.

hep-ph↗

arXiv2026

Mixture-of-Thought-Tokens: Unifying Perception and Reasoning for Free-form Multimodal Grounding

Multimodal Large Language Models have made great progress in grounding tasks, yet existing methods still struggle to unify precise localization and complex reasoning. For one thing, text-based methods rely on coordinates or index prediction, severely limiting the perceptual capabilities of the model for dense visual objects. Meanwhile, latent token-based methods employ special tokens without inherent spatial references and use a decoding mechanism that lacks thinking steps, weakening high-level reasoning capabilities. Consequently, developing a unified framework that excels in both perception and reasoning remains challenging. To address this, we propose Mixture-of-Thought-Tokens (Motto), a new free-form multimodal grounding method that bridges the perception-reasoning gap, enabling MLLMs to empower diverse, arbitrary grounding queries. Specifically, we introduce Spatially-Grounded Thought Tokenization to explicitly align special tokens with spatial locations for clear spatial correspondence and visual interpretability. We further design a Context-Adaptive Chain-of-Tokens that dynamically switch grounding modes within an interleaved reasoning chain, achieving robust grounding across tasks of varying complexity. In addition, we construct PR-Bench, a new referring expression comprehension benchmark to evaluate the perception-reasoning gap. Extensive experiments demonstrate that Motto achieves state-of-the-art performance across diverse free-form grounding tasks.

cs.CV↗

arXiv2025

Balanced Collaborative Exploration via Distributed Topological Graph Voronoi Partition

This work addresses the collaborative multi-robot autonomous online exploration problem, particularly focusing on distributed exploration planning for dynamically balanced exploration area partition and task allocation among a team of mobile robots operating in obstacle-dense non-convex environments. We present a novel topological map structure that simultaneously characterizes both spatial connectivity and global exploration completeness of the environment. The topological map is updated incrementally to utilize known spatial information for updating reachable spaces, while exploration targets are planned in a receding horizon fashion under global coverage guidance. A distributed weighted topological graph Voronoi algorithm is introduced implementing balanced graph space partitions of the fused topological maps. Theoretical guarantees are provided for distributed consensus convergence and equitable graph space partitions with constant bounds. A local planner optimizes the visitation sequence of exploration targets within the balanced partitioned graph space to minimize travel distance, while generating safe, smooth, and dynamically feasible motion trajectories. Comprehensive benchmarking against state-of-the-art methods demonstrates significant improvements in exploration efficiency, completeness, and workload balance across the robot team.

cs.RO↗

arXiv2025

Adaptive Gate-Aware Mamba Networks for Magnetic Resonance Fingerprinting

Magnetic Resonance Fingerprinting (MRF) enables fast quantitative imaging by matching signal evolutions to a predefined dictionary. However, conventional dictionary matching suffers from exponential growth in computational cost and memory usage as the number of parameters increases, limiting its scalability to multi-parametric mapping. To address this, recent work has explored deep learning-based approaches as alternatives to DM. We propose GAST-Mamba, an end-to-end framework that combines a dual Mamba-based encoder with a Gate-Aware Spatial-Temporal (GAST) processor. Built on structured state-space models, our architecture efficiently captures long-range spatial dependencies with linear complexity. On 5 times accelerated simulated MRF data (200 frames), GAST-Mamba achieved a T1 PSNR of 33.12~dB, outperforming SCQ (31.69~dB). For T2 mapping, it reached a PSNR of 30.62~dB and SSIM of 0.9124. In vivo experiments further demonstrated improved anatomical detail and reduced artifacts. Ablation studies confirmed that each component contributes to performance, with the GAST module being particularly important under strong undersampling. These results demonstrate the effectiveness of GAST-Mamba for accurate and robust reconstruction from highly undersampled MRF acquisitions, offering a scalable alternative to traditional DM-based methods.

eess.IV↗

arXiv2025

Illuminating Very Heavy Dark Matter in the Earth with Tau Neutrinos

Dark matter accumulates in the center of the Earth as the planet plows through the dark matter halo in the Milky Way. Possible annihilation of dark matter to Standard Model particles can be probed in indirect dark matter searches. Among possible messengers, neutrinos are uniquely ideal as they can escape dense regions. Therefore, neutrino telescopes, with their large volume and broad energy exposures, offer new opportunities to search for dark matter signals from the center of the Earth. However, such studies have been restricted to dark matter masses below $\sim$ PeV as the Earth becomes opaque to very-high-energy neutrinos. In this study, we demonstrate that neutrino telescopes operating at TeV-PeV energies can probe very heavy dark matter particles if they annihilate to tau neutrinos or tau leptons. Here, we report upper limits on the spin-independent dark matter-nucleon cross section for masses between $10^5$ GeV and $10^{10}$ GeV by using 7.5 years of IceCube high-energy starting event observations. Our results motivate detailed analyses in IceCube and other upcoming neutrino telescopes in the Northern Hemisphere.

hep-ph↗

arXiv2024

Resource-Efficient Cooperative Online Scalar Field Mapping via Distributed Sparse Gaussian Process Regression

Cooperative online scalar field mapping is an important task for multi-robot systems. Gaussian process regression is widely used to construct a map that represents spatial information with confidence intervals. However, it is difficult to handle cooperative online mapping tasks because of its high computation and communication costs. This letter proposes a resource-efficient cooperative online field mapping method via distributed sparse Gaussian process regression. A novel distributed online Gaussian process evaluation method is developed such that robots can cooperatively evaluate and find observations of sufficient global utility to reduce computation. The bounded errors of distributed aggregation results are guaranteed theoretically, and the performances of the proposed algorithms are validated by real online light field mapping experiments.

cs.RO↗

arXiv2023

Entanglement distillation in terms of Schmidt rank and matrix rank

Entanglement distillation is a key task in quantum-information processing. In this paper, we distill non-positive-partial-transpose (NPT) bipartite states of some given Schmidt rank and matrix rank. We show that all bipartite states of Schmidt rank two are locally equivalent to classical-classical states, and all bipartite states of Schmidt rank three are 1-undistillable. Subsequently, we show that low-rank B-irreducible NPT states are distillable for large-rank reduced density operators by proving low-rank B-irreducible NPT state whose range contains a product vector is distillable. Eventually, we present an equivalent condition to distill $M\times N$ bipartite states of rank $\max\{M,N\}+1$.

quant-ph↗