arXiv subjects

Haoyang Liu

Publications and source records attributed to Haoyang Liu.

At least 37 records · Page 2Linked to original sources

arXiv2026

Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks

The Sagittarius (Sgr) Stream is produced by the ongoing disruption of the Sgr dwarf spheroidal (dSph) galaxy and is thought to contain multiple wraps that were stripped during different pericentric passages. In this study, we introduce a neural-network--based method trained on $N$-body simulations to infer the stripping time of Sgr Stream stars directly from their phase-space coordinates. We combine spectroscopic data from SEGUE, APOGEE DR17, and LAMOST DR7 LRS with \textit{Gaia} EDR3 astrometry and distance estimates from the latest \texttt{StarHorse} catalog to identify high-quality Sgr Stream members. Applying our method to these stars, we measure a clear metallicity gradient with stripping time, well described by a linear relation with slope $\sim 0.3~\mathrm{dex~Gyr^{-1}}$. We further predict the stripping times of globular clusters previously suggested to originate from the Sgr dSph. M 54, Terzan 7, Terzan 8, and Arp 2 exhibit stripping times consistent with being currently bound to the Sgr remnant. Pal 12, Whiting 1, and NGC 2419 are inferred to have been stripped $0.9 \pm 0.1$, $1.1 \pm 0.2$, and $2.1 \pm 0.2$ Gyr ago, respectively. For NGC 4147 and NGC 5634, whose membership in the Sgr system remains uncertain, our analysis suggests stripping times of $1.1 \pm 0.4$ and $1.1 \pm 0.1$ Gyr, respectively, if they are ultimately confirmed as genuine Sgr members. These results demonstrate that data-driven models of dynamical stripping histories offer a promising approach for reconstructing the formation and chemical evolution of the Sgr Stream.

astro-ph.GA↗

arXiv2026

Rethinking Entropy Interventions in RLVR: An Entropy Change Perspective

Reinforcement Learning with Verifiable Rewards (RLVR) serves as a cornerstone technique for enhancing the reasoning capabilities of Large Language Models (LLMs). However, its training is often plagued by \emph{entropy collapse}, a rapid decline in policy entropy that limits exploration and undermines training effectiveness. While recent works attempt to mitigate this issue via several heuristic entropy interventions, the underlying mechanisms remain poorly understood. In this work, we conduct comprehensive theoretical and empirical analyses of entropy dynamics in RLVR, offering two main insights: (1) We derive a tight analytical approximation for token-level entropy change at each update step, revealing four governing factors and providing a unified theoretical framework to explain how existing methods influence entropy; (2) We reveal a fundamental limitation of recent approaches: they rely on heuristic adjustments to one or two of these factors, leaving other relevant factors unconsidered, thus inherently limiting their effectiveness. Motivated by these findings, we propose STEER, a principled entropy-modulation method that adaptively reweights tokens based on theoretically-estimated entropy variations. Extensive experiments across six mathematical reasoning and three coding benchmarks demonstrate that STEER effectively mitigates entropy collapse and consistently outperforms state-of-the-art baselines.

cs.LG↗

arXiv2026

Joint Optimization of Multi-agent Memory System

Memory systems are critical for LLMs, mitigating context window limitations and supporting long-horizon user-LLM interactions. Such systems typically comprise multiple agents responsible for memory construction and retrieval. Existing approaches often optimize each agent independently under a shared global objective (e.g., downstream QA accuracy), treating other agents as a static environment. However, this design has two key limitations: (1) independent optimization ignores inter-agent dependencies and lacks agents' co-adaptation, and (2) relying solely on sparse global rewards provides limited guidance for optimizing specialized agents and causes ambiguous credit assignment. These may ultimately limit agents' effective collaboration in the memory system. To address these limitations, we propose CoMAM, a joint optimization framework that promotes collaboration among agents via end-to-end reinforcement learning and an adaptive credit assignment mechanism. Specifically, we model the multi-agent pipeline as a Markov decision process (MDP) to expose inter-agent dependencies during end-to-end training. Agents are then jointly optimized using a combination of their local task reward and an adaptively weighted global reward, enabling agents to co-adapt while receiving targeted feedback for their respective roles. Experiments show that CoMAM consistently outperforms leading memory systems, validating the effectiveness of the joint optimization framework.

cs.MA↗

arXiv2026

NED-Tree: Bridging the Semantic Gap with Nonlinear Element Decomposition Tree for LLM Nonlinear Optimization Modeling

Automating the translation of Operations Research (OR) problems from natural language to executable models is a critical challenge. While Large Language Models (LLMs) have shown promise in linear tasks, they suffer from severe performance degradation in real-world nonlinear scenarios due to semantic misalignment between mathematical formulations and solver codes, as well as unstable information extraction. In this study, we introduce NED-Tree, a systematic framework designed to bridge the semantic gap. NED-Tree employs (a) a sentence-by-sentence extraction strategy to ensure robust parameter mapping and traceability; and (b) a recursive tree-based structure that adaptively decomposes complex nonlinear terms into solver-compatible sub-elements. Additionally, we present NEXTOR, a novel benchmark specifically designed for complex nonlinear, extensive-constraint OR problems. Experiments across 10 benchmarks demonstrate that NED-Tree establishes a new state-of-the-art with 72.51% average accuracy, NED-Tree is the first framework that drives LLMs to resolve nonlinear modeling difficulties through element decomposition, achieving alignment between modeling semantics and code semantics. The NED-Tree framework and benchmark are accessible in the anonymous repository https://anonymous.4open.science/r/NORA-NEXTOR.

cs.AI↗

arXiv2026

Revisiting candidate high-velocity stars associated with the Sagittarius dwarf spheroidal galaxy

Hypervelocity stars (HVSs) are valuable tracers of extreme dynamical processes. The Sagittarius dwarf spheroidal galaxy (Sgr dSph), currently undergoing tidal disruption, offers a unique environment to search for such stars. We aim to identify candidate HVSs dynamically linked to the Sgr dSph and to assess their possible origins. Using Gaia DR3, DESI DR1, and LAMOST DR12, we selected stars with galactocentric velocities above 400 km\,s$^{-1}$ and traced their orbits in a realistic Galactic potential including the Sgr dSph and the Large Magellanic Cloud. We then tested three scenarios for their origin: the Hills mechanism, tidal disruption, and random halo star encounters. We identified 95 candidates passing within 2.5 half-mass radii of the Sgr dSph. Their kinematics are inconsistent with production by the Hills mechanism or tidal disruption but are well reproduced by halo stars that naturally cross the Sgr orbit. Furthermore, their metallicity distribution is consistent with that of the Milky Way halo rather than the Sgr stream or Sgr dSph. Our results suggest that our candidates and those in previous studies are most likely halo stars rather than genuine Sgr-origin HVSs. This highlights the need to account for the halo population when inferring stellar origins from orbital analysis and that chemical abundances will be a valuable constraint in the future. While we detect no unbound Sgr HVSs, such a discovery would directly imply extreme dynamical processes. Our results serve as a basis for future studies with upcoming surveys.

astro-ph.GA↗

arXiv2026

RobustSCI: Beyond Reconstruction to Restoration for Snapshot Compressive Imaging under Real-World Degradations

Deep learning algorithms for video Snapshot Compressive Imaging (SCI) have achieved great success, yet they predominantly focus on reconstructing from clean measurements. This overlooks a critical real-world challenge: the captured signal itself is often severely degraded by motion blur and low light. Consequently, existing models falter in practical applications. To break this limitation, we pioneer the first study on robust video SCI restoration, shifting the goal from "reconstruction" to "restoration"--recovering the underlying pristine scene from a degraded measurement. To facilitate this new task, we first construct a large-scale benchmark by simulating realistic, continuous degradations on the DAVIS 2017 dataset. Second, we propose RobustSCI, a network that enhances a strong encoder-decoder backbone with a novel RobustCFormer block. This block introduces two parallel branches--a multi-scale deblur branch and a frequency enhancement branch--to explicitly disentangle and remove degradations during the recovery process. Furthermore, we introduce RobustSCI-C (RobustSCI-Cascade), which integrates a pre-trained Lightweight Post-processing Deblurring Network to significantly boost restoration performance with minimal overhead. Extensive experiments demonstrate that our methods outperform all SOTA models on the new degraded testbeds, with additional validation on real-world degraded SCI data confirming their practical effectiveness, elevating SCI from merely reconstructing what is captured to restoring what truly happened.

cs.CV↗

arXiv2026

Chemical Properties and Sagittarius-induced Dynamical Perturbations of the GD-1 Stream

In this study, we investigate the chemical properties of the GD-1 stream using cross-matched, data-driven elemental abundances. The results reveal no clear $α$-knee in the [Mg/Fe]-[Fe/H] plane, and strong abundance consistency between the thin stream and cocoon, supporting a common origin. The absence of multiple-population signatures (e.g., C-N anti-correlation) suggests a low-mass progenitor. Using a test-particle simulation with the particle spray method and including perturbations from the Sagittarius (Sgr) dwarf galaxy, it shows that Sgr does not significantly heat the stream to form the cocoon, but modifies the intrinsic $ϕ_2$ distribution, in agreement with observations. The trailing arm narrowly distributed across the width of the stream, while the leading arm is more diffuse, indicating that major fraction of cocoon stars are present towards the leading arm. Sgr also drags more stream particles moving toward the Galactic center, producing an excess at $V_{\text{GSR}}<0$, consistent with data. Our study confirms the Sgr has a non-negligible dynamical influence on the GD-1 stream. Other heating mechanisms (e.g., dark matter sub-halo encounters and pre-stripping process inside the parent halo) remain to be considered, and higher-resolution spectroscopy is needed to further constrain chemical abundances.

astro-ph.GA↗

arXiv2026

Tunable interplay of orbital and spin magnetization in trigonal tellurium

Orbital effects, despite their fundamental significance and potential to engender novel physical phenomena and enable new applications, have long been underexplored compared to their spin counterparts. Recently, surging interest in the orbital degree of freedom has led to the discovery of a plethora of orbital-related effects, underscoring the need for a deeper understanding of their roles in quantum materials. Here, we report systematic experimental evidence consistent with orbital magnetization and spontaneous rotational symmetry breaking in trigonal Tellurium, an elemental semiconductor with a unique helical crystal structure that serves as a natural platform for investigating orbital effects. Detailed angular dependent linear and nonlinear magnetotransport measurements, supported by symmetry-guided Boltzmann transport analysis, support the interpretation of coexistence of current-induced spin polarization and orbital magnetization. With the goal of disentangling the interplay between spin and orbital degrees of freedom through electrostatic gating, this work establishes a general framework for understanding orbital magnetization in chiral crystals and beyond, paving the way for its utilization in orbitronics and spintronics.

cond-mat.mtrl-sci↗

arXiv2026

PPI-SVRG: Unifying Prediction-Powered Inference and Variance Reduction for Semi-Supervised Optimization

We study semi-supervised stochastic optimization when labeled data is scarce but predictions from pre-trained models are available. PPI and SVRG both reduce variance through control variates -- PPI uses predictions, SVRG uses reference gradients. We show they are mathematically equivalent and develop PPI-SVRG, which combines both. Our convergence bound decomposes into the standard SVRG rate plus an error floor from prediction uncertainty. The rate depends only on loss geometry; predictions affect only the neighborhood size. When predictions are perfect, we recover SVRG exactly. When predictions degrade, convergence remains stable but reaches a larger neighborhood. Experiments confirm the theory: PPI-SVRG reduces MSE by 43--52\% under label scarcity on mean estimation benchmarks and improves test accuracy by 2.7--2.9 percentage points on MNIST with only 10\% labeled data.

cs.LG↗

arXiv2026

A Low-mass Model of The Milky Way: The Disk Warp Resulting from A Galaxy Merger

Previous studies have shown that disk warps can result from galaxy mergers. Recent research indicates a noticeable decline in the rotation curve (RC) of the Milky Way (MW), suggesting the need for a new low-mass model to describe its dynamical features. This study constructs a new Gaia-Sausage-Enceladus (GSE) merger model to characterize the RC features of our galaxy. We use the GIZMO code to simulate mergers with various orbital parameters to investigate how the disk warp evolves under different conditions. This simulation demonstrates the evolutionary mechanism of disk warp, which arises due to the asymmetric gravitational potential of the dark matter (DM) halo generated universally by galaxy mergers. The results indicate that the tilt angle of the DM halo partly reflects the gravitational strength at the $Z=0$ plane, while the gravitational strength on the disk plane reflects the amplitude of disk warp. We identify a dual-regime interaction mechanism driven by the asymmetric halo potential. On short timescales, we find a distinct anti-correlation between the halo's tilt angle and the disk's warp amplitude, indicating a `seesaw' mechanism of angular momentum exchange. On secular timescales, however, dynamical friction drives a global alignment, causing both the halo tilt and the warp amplitude to decay simultaneously. Furthermore, we demonstrate that high-inclination mergers can sustain long-lived prograde precession, where the persistent yet decaying gravitational torque maintains the prograde bending mode against differential wind-up.

astro-ph.GA↗

arXiv2026

Bi-Mem: Bidirectional Construction of Hierarchical Memory for Personalized LLMs via Inductive-Reflective Agents

Constructing memory from users' long-term conversations overcomes LLMs' contextual limitations and enables personalized interactions. Recent studies focus on hierarchical memory to model users' multi-granular behavioral patterns via clustering and aggregating historical conversations. However, conversational noise and memory hallucinations can be amplified during clustering, causing locally aggregated memories to misalign with the user's global persona. To mitigate this issue, we propose Bi-Mem, an agentic framework ensuring hierarchical memory fidelity through bidirectional construction. Specifically, we deploy an inductive agent to form the hierarchical memory: it extracts factual information from raw conversations to form fact-level memory, aggregates them into thematic scenes (i.e., local scene-level memory) using graph clustering, and infers users' profiles as global persona-level memory. Simultaneously, a reflective agent is designed to calibrate local scene-level memories using global constraints derived from the persona-level memory, thereby enforcing global-local alignment. For coherent memory recall, we propose an associative retrieval mechanism: beyond initial hierarchical search, a spreading activation process allows facts to evoke contextual scenes, while scene-level matches retrieve salient supporting factual information. Empirical evaluations demonstrate that Bi-Mem achieves significant improvements in question answering performance on long-term personalized conversational tasks.

cs.MA↗

arXiv2026

A reconstructed discontinuous approximation for distributed elliptic control problems

In this paper, we present and analyze an interior penalty discontinuous Galerkin method for the distributed elliptic optimal control problems. It is based on a reconstructed discontinuous approximation which admits arbitrarily high-order approximation space with only one unknown per element. Applying this method, we develop a proper discretization scheme that approximates the state and adjoint variables in the approximation space. Our main contributions are twofold: (1) the derivation of both a priori and a posteriori error estimates of the $L^2$-norm and the energy norms, and (2) the implementation of an efficiently solvable discrete system, which is solved via a linearly convergent projected gradient descent method. Numerical experiments are provided to verify the convergence order in a priori error estimate and the efficiency of a posteriori error estimate.

math.NA↗

arXiv2025

SymMaP: Improving Computational Efficiency in Linear Solvers through Symbolic Preconditioning

Matrix preconditioning is a critical technique to accelerate the solution of linear systems, where performance heavily depends on the selection of preconditioning parameters. Traditional parameter selection approaches often define fixed constants for specific scenarios. However, they rely on domain expertise and fail to consider the instance-wise features for individual problems, limiting their performance. In contrast, machine learning (ML) approaches, though promising, are hindered by high inference costs and limited interpretability. To combine the strengths of both approaches, we propose a symbolic discovery framework-namely, Symbolic Matrix Preconditioning (SymMaP)-to learn efficient symbolic expressions for preconditioning parameters. Specifically, we employ a neural network to search the high-dimensional discrete space for expressions that can accurately predict the optimal parameters. The learned expression allows for high inference efficiency and excellent interpretability (expressed in concise symbolic formulas), making it simple and reliable for deployment. Experimental results show that SymMaP consistently outperforms traditional strategies across various benchmarks.

math.NA↗

arXiv2025

CoCo-MILP: Inter-Variable Contrastive and Intra-Constraint Competitive MILP Solution Prediction

Mixed-Integer Linear Programming (MILP) is a cornerstone of combinatorial optimization, yet solving large-scale instances remains a significant computational challenge. Recently, Graph Neural Networks (GNNs) have shown promise in accelerating MILP solvers by predicting high-quality solutions. However, we identify that existing methods misalign with the intrinsic structure of MILP problems at two levels. At the leaning objective level, the Binary Cross-Entropy (BCE) loss treats variables independently, neglecting their relative priority and yielding plausible logits. At the model architecture level, standard GNN message passing inherently smooths the representations across variables, missing the natural competitive relationships within constraints. To address these challenges, we propose CoCo-MILP, which explicitly models inter-variable Contrast and intra-constraint Competition for advanced MILP solution prediction. At the objective level, CoCo-MILP introduces the Inter-Variable Contrastive Loss (VCL), which explicitly maximizes the embedding margin between variables assigned one versus zero. At the architectural level, we design an Intra-Constraint Competitive GNN layer that, instead of homogenizing features, learns to differentiate representations of competing variables within a constraint, capturing their exclusionary nature. Experimental results on standard benchmarks demonstrate that CoCo-MILP significantly outperforms existing learning-based approaches, reducing the solution gap by up to 68.12% compared to traditional solvers. Our code is available at https://github.com/happypu326/CoCo-MILP.

cs.LG↗

arXiv2025

RoME: Domain-Robust Mixture-of-Experts for MILP Solution Prediction across Domains

Mixed-Integer Linear Programming (MILP) is a fundamental and powerful framework for modeling complex optimization problems across diverse domains. Recently, learning-based methods have shown great promise in accelerating MILP solvers by predicting high-quality solutions. However, most existing approaches are developed and evaluated in single-domain settings, limiting their ability to generalize to unseen problem distributions. This limitation poses a major obstacle to building scalable and general-purpose learning-based solvers. To address this challenge, we introduce RoME, a domain-Robust Mixture-of-Experts framework for predicting MILP solutions across domains. RoME dynamically routes problem instances to specialized experts based on learned task embeddings. The model is trained using a two-level distributionally robust optimization strategy: inter-domain to mitigate global shifts across domains, and intra-domain to enhance local robustness by introducing perturbations on task embeddings. We reveal that cross-domain training not only enhances the model's generalization capability to unseen domains but also improves performance within each individual domain by encouraging the model to capture more general intrinsic combinatorial patterns. Specifically, a single RoME model trained on three domains achieves an average improvement of 67.7% then evaluated on five diverse domains. We further test the pretrained model on MIPLIB in a zero-shot setting, demonstrating its ability to deliver measurable performance gains on challenging real-world instances where existing learning-based approaches often struggle to generalize.

cs.LG↗

arXiv2025

Kinematic, Elemental and Structural Dependences on Metallicity in the Galactic Bulge

We selected Bulge stars from APOGEE DR17 cross-matched with astrometric data from \textit{Gaia} DR3. Bulge stars were divided into sub-samples with line-of-sight velocity dispersion analyzed and the peaks of MDF were detected by both Gaussian Mixture Models (GMM) and \texttt{scipy.signal.find\_peaks}. GMM is also conducted to kinematically distinguish the metal-poor and metal-rich populations. Analyses were put on the Bulge stars (including retrograde stars), their elemental abundances, and the [Mg/Mn]-[Al/Fe] plane to investigate potential accreted components. Finally, the shapes (X-shaped/boxy) of Bulge stars with different metallicities were analyzed through least-squares fitting based on the analytical Bulge models. By studying the kinematic, elemental and structural dependences on metallicity for Bulge stars, our findings are concluded as follows: 1. Six peaks are detected in the Bulge MDF, encompassing values reported in previous studies, suggesting a complex composition of Bulge populations. 2. An inversion relationship is well-observed in metal-rich sub-samples, while absent in metal-poor sub-samples. 3. Metal-poor populations exhibit larger dispersions than metal-rich stars (which is also revealed by GMM decomposition), suggesting that metal-rich stars are kinematically coherent. 4. Retrograde stars are confined to $\sim1$ kpc of the Galactic center, with their relative fraction decreasing at higher [Fe/H] -- a trend potentially linked to the ``spin-up'' process of Galactic disks. 5. Metal-rich Bulge stars with [Al/Fe] $<-0.15$ are likely associated with from disk accreted substructure, while all elemental planes exhibit bimodality but Na abundances rise monotonically with metallicity. 6. In general, stars with all metallicities support a boxy profile.

astro-ph.GA↗

arXiv2025

OptiTree: Hierarchical Thoughts Generation with Tree Search for LLM Optimization Modeling

Optimization modeling is one of the most crucial but technical parts of operations research (OR). To automate the modeling process, existing works have leveraged large language models (LLMs), prompting them to break down tasks into steps for generating variables, constraints, and objectives. However, due to the highly complex mathematical structures inherent in OR problems, standard fixed-step decomposition often fails to achieve high performance. To address this challenge, we introduce OptiTree, a novel tree search approach designed to enhance modeling capabilities for complex problems through adaptive problem decomposition into simpler subproblems. Specifically, we develop a modeling tree that organizes a wide range of OR problems based on their hierarchical problem taxonomy and complexity, with each node representing a problem category and containing relevant high-level modeling thoughts. Given a problem to model, we recurrently search the tree to identify a series of simpler subproblems and synthesize the global modeling thoughts by adaptively integrating the hierarchical thoughts. Experiments show that OptiTree significantly improves the modeling accuracy compared to the state-of-the-art, achieving over 10\% improvements on the challenging benchmarks. The code is released at https://github.com/MIRALab-USTC/OptiTree/tree/main.

cs.AI↗

arXiv2025

Just-In-Time Piecewise-Linear Semantics for ReLU-type Networks

We present a JIT PL semantics for ReLU-type networks that compiles models into a guarded CPWL transducer with shared guards. The system adds hyperplanes only when operands are affine on the current cell, maintains global lower/upper envelopes, and uses a budgeted branch-and-bound. We obtain anytime soundness, exactness on fully refined cells, monotone progress, guard-linear complexity (avoiding global $\binom{k}{2}$), dominance pruning, and decidability under finite refinement. The shared carrier supports region extraction, decision complexes, Jacobians, exact/certified Lipschitz, LP/SOCP robustness, and maximal causal influence. A minimal prototype returns certificates or counterexamples with cost proportional to visited subdomains.

cs.LO↗