arXiv subjects

Yucheng Yang

Publications and source records attributed to Yucheng Yang.

At least 19 recordsLinked to original sources

arXiv2026

Synergy between laser linewidth and frequency chirp in mesospheric magnetometry based on the sodium laser guide star

Mesospheric sodium magnetometry with a laser guide star measures the geomagnetic field near 90~km. Its sensitivity hinges on laser linewidth and chirp, yet prior work optimized these two parameters only separately. We use velocity-resolved density-matrix simulations of Larmor-synchronous pulsed Na D$_2$ pumping to scan both parameters jointly. Linewidth and chirp exhibit a synergy: when chirping carries the recoil-mitigation role, the return flux stays within 1\% of its peak across linewidths of 2--10~MHz. The flux-optimal chirp is $0.19~\mathrm{MHz/\upmu s}$, one fifth of the continuous-wave rate; this synergy guides high-sensitivity mesospheric magnetometer design.

physics.app-ph↗

arXiv2026

Which transition can be used in sodium mirrorless lasing for mesospheric magnetometry?

Directed mirrorless lasing from the mesospheric sodium layer has been proposed as a way to overcome the isotropy of laser-guide-star fluorescence, with demonstrated cell-scale analogues and a demonstrated stand-off magnetometry application. Several transition paths on the Na ladder compete for the same pump photons, and which of them can sustain a population inversion---and under what pumping format---is usually settled by numerical scans of a multi-level rate model. Here the same question is answered in closed form, and three design-relevant results follow from the atomic data alone. (1) A cascade inversion on $u\to l$ requires $\tau_u/\tau_l>(g_u/g_l)b_{ul}$. The inequality reproduces the full continuous-wave classification---admitting $4P_{3/2}\to4S_{1/2}$ (2.21~\textmu m), $4P_{3/2}\to3D_{5/2}$ (9.1~\textmu m) and the fine-structure companion $4S_{1/2}\to3P_{1/2}$ (1138~nm), excluding $4D_{5/2}\to4P_{3/2}$ (2.34~\textmu m) and $3D_{5/2}\to3P_{3/2}$ (819~nm)---and shows the 2.34~\textmu m line to miss by only 9\%, which is why it is transiently accessible. (2) Velocity selectivity reverses the naive scheme ranking because a Doppler-averaged treatment, exact for one-step pumping, underestimates two-step pumping by a participation factor of order the Doppler-to-natural width ratio, $\sim$$10^{2}$; equal division of power between two pump beams is exactly optimal. (3) The transient window on 2.34~\textmu m closes after $\sim$2$\tau(4P)\approx210$~ns, set by the reservoir lifetime rather than by the pump. Each conclusion is traceable to a lifetime, a branching ratio or a linewidth, and transfers to another species without rerunning a model.

physics.app-ph↗

arXiv2026

Optimization of the Repumping Parameters for a Sodium Laser Guide Star Magnetometer

A sodium laser guide star operated as a mesospheric magnetometer modulates a 589 nm laser at the local Larmor frequency and usually diverts a fraction of its power to a repumping light that recovers atoms lost to the dark ground state.The polarization, read out for the most strongly driven velocity group, calls for 2.8 times the flux optimal fraction, and a shot noise figure of merit combining the two observables for 2 times, beyond the range commercial guide star lasers provide.

physics.app-ph↗

arXiv2026

Population-inversion map of the mesospheric sodium ladder: continuous-wave and pulsed pumping schemes for directed emission

Directed mirrorless lasing from the mesospheric sodium layer has been proposed as a way to overcome the isotropy of laser guide star fluorescence, with demonstrated cell-scale analogues and a demonstrated stand-off magnetometry application. Several transition paths on the Na ladder compete for the same pump photons. We build a ten-level rate-equation model of the ladder from NIST transition probabilities and evaluate every electric-dipole line under four continuous-wave pumping schemes, both in a Doppler-averaged treatment and in a velocity-selective treatment appropriate for the collision-poor mesosphere. Three design-relevant results emerge. At practically accessible continuous-wave irradiances the column-gain exponents remain far below unity, consistent with published feasibility estimates; the value of the classification is to identify which lines, schemes, and pulse formats merit further study.

physics.optics↗

arXiv2026

Algorithm-Driven SVARs: Navigating the Wilderness of Big Data

Every SVAR result is conditional on two choices: the restrictions that identify the shock and the variables on which they operate. The literature disciplines the first; the second is chosen by hand. We develop a Bayesian methodology that constructs information sets, uses an out-of-sample criterion, and retains the largest system it admits. Under recursive identification, output rises with housing production rather than household credit alone. For monetary policy, an anchor-free joint Bayesian proxy SVAR with multiple instruments strengthens the credit spread channel. A core system augmented with the selected corporate spread identifies expected default risk as a potent transmission margin.

econ.EM↗

arXiv2026

GuidedBridge: Training-freely Improving Bridge Models with Prior Guidance

Guidance methods, such as classifier-free guidance (CFG) and auto-guidance (AG), have advanced noise-to-data generation in diffusion models. Recently, bridge models have introduced a data-to-data generative process that can exploit an instructive clean prior. In this work, inspired by previous methods creating quality difference between denoising results as guidance, we propose a training-free bridge guidance method, termed Prior Guidance (PG). Specifically, we introduce a weak prior, which is unseen during bridge pre-training, hindering prior exploitation and thereby degrading denoising result. Then, we contrast it with the seen prior to highlight and enhance prior exploitation via a scaling factor. Moreover, we analyze the underlying mechanism of prior exploitation in the bridge process and design frequency-modulated prior guidance (FMPG), which tailors the guidance scale to low- and high-frequency bands coherent with bridge generative dynamics. To address prior exploitation in image in-painting, we develop a cascaded framework, CFG-FMPG, which first generates a noisy hidden representation via CFG and then exploits it as a generative prior with FMPG, fulfilling their complementary strengths without compromising inference efficiency. Experiments demonstrate that our PG methods consistently improve pre-trained bridge models across diverse image translation tasks.

cs.CV↗

arXiv2026

MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI

Modern AI progress has been driven by ML methods that are generalizable across settings and scalable to larger regimes. As large language models demonstrate advanced capabilities in reasoning, coding, and engineering tasks, it is increasingly important to understand whether they can discover such methods rather than only apply existing ones. We introduce MLS-Bench, a benchmark for evaluating whether AI systems can invent generalizable and scalable ML methods. MLS-Bench contains 140 tasks across 12 domains, each requiring an agent to improve one targeted component of an ML system or algorithm and demonstrate that the improvement generalizes across controlled settings and scales. We find that current agents remain far from reliably surpassing human-designed methods, and that engineering-style tuning is easier for them than genuine method invention. We further study the effects of test-time scaling, adaptive compute allocation, and context provision on agents' discovery performance, together with case studies of their behavior. Our analyses suggest that the bottleneck is not only in proposing new methods, but also in the scientific insight needed to plan, validate, and scale claims about them. More search, compute, or context alone does not remove this bottleneck. We build and maintain a community platform for cumulative and comparable iteration, and release the data and code at https://mls-bench.com.

cs.LG↗

arXiv2026

One Model for All: Multi-Objective Controllable Language Models

Aligning large language models (LLMs) with human preferences is critical for enhancing LLMs' safety, helpfulness, humor, faithfulness, etc. Current reinforcement learning from human feedback (RLHF) mainly focuses on a fixed reward learned from average human ratings, which may weaken the adaptability and controllability of varying preferences. However, creating personalized LLMs requires aligning LLMs with individual human preferences, which is non-trivial due to the scarce data per user and the diversity of user preferences in multi-objective trade-offs, varying from emphasizing empathy in certain contexts to demanding efficiency and precision in others. Can we train one LLM to produce personalized outputs across different user preferences on the Pareto front? In this paper, we introduce Multi-Objective Control (MOC), which trains a single LLM to directly generate responses in the preference-defined regions of the Pareto front. Our approach introduces multi-objective optimization (MOO) principles into RLHF to train an LLM as a preference-conditioned policy network. We improve the computational efficiency of MOC by applying MOO at the policy level, enabling us to fine-tune a 7B-parameter model on a single A6000 GPU. Extensive experiments demonstrate the advantages of MOC over baselines in three aspects: (i) controllability of LLM outputs w.r.t. user preferences on the trade-off among multiple rewards; (ii) quality and diversity of LLM outputs, measured by the hyper-volume of multiple solutions achieved; and (iii) generalization to unseen preferences. These results highlight MOC's potential for real-world applications requiring scalable and customizable LLMs.

cs.LG↗

arXiv2026

Recurrent Structural Policy Gradient for Partially Observable Mean Field Games

Mean Field Games (MFGs) provide a principled framework for modelling interactions in large population systems. However, algorithmic progress has been limited since model-free methods are high variance and exact methods scale poorly. Recent Hybrid Structural Methods (HSMs) reduce variance while maintaining tractability by leveraging low-dimensional individual state and action spaces and known transition dynamics to compute the exact expected return conditioned on Monte Carlo rollouts of common noise. However, HSMs have not been extended to partially observable settings. We propose Recurrent Structural Policy Gradient (RSPG), the first history-aware HSM for MFGs with public partial information. RSPG achieves an order-of-magnitude faster convergence than model-free RL methods while learning history-aware behaviour, unlike current HSMs. To facilitate research into MFGs, we also introduce MFAX, our JAX-based framework for MFGs that supports both analytic and sample-based mean-field updates. MFAX and usage examples can be found at https://clarisse-wibault.github.io/rspg/.

cs.AI↗

arXiv2025

Deep Learning for Art Market Valuation

We study how deep learning can improve valuation in the art market by incorporating the visual content of artworks into predictive models. Using a large repeated-sales dataset from major auction houses, we benchmark classical hedonic regressions and tree-based methods against modern deep architectures, including multi-modal models that fuse tabular and image data. We find that while artist identity and prior transaction history dominate overall predictive power, visual embeddings provide a distinct and economically meaningful contribution for fresh-to-market works where historical anchors are absent. Interpretability analyses using Grad-CAM and embedding visualizations show that models attend to compositional and stylistic cues. Our findings demonstrate that multi-modal deep learning delivers significant value precisely when valuation is hardest, namely first-time sales, and thus offers new insights for both academic research and practice in art market valuation.

q-fin.GN↗

arXiv2025

Enabling Ultra-Fast Cardiovascular Imaging Across Heterogeneous Clinical Environments with A Generalist Foundation Model and Multimodal Database

Multimodal cardiovascular magnetic resonance (CMR) imaging provides comprehensive and non-invasive insights into cardiovascular disease (CVD) diagnosis and underlying mechanisms. Despite decades of advancements, its widespread clinical adoption remains constrained by prolonged scan times, inconsistent image quality, and heterogeneity across medical environments. This underscores the urgent need for a generalist reconstruction foundation model for ultra-fast CMR imaging, one formulated for physics-constrained inverse problems in the sensor (k-space) domain, capable of adapting across diverse imaging scenarios and serving as the essential substrate for all downstream analyses. To enable this goal, we curate MMCMR-427K, the largest and most comprehensive multimodal CMR k-space database to date, comprising 427,465 multi-coil k-space data paired with structured metadata across 13 international centers, 12 CMR modalities, 15 scanners spanning four field strengths, and 17 CVD categories in populations across three continents. Building on this unprecedented resource, we introduce CardioMM, a generalist reconstruction foundation model capable of dynamically adapting to heterogeneous fast CMR imaging scenarios. CardioMM unifies semantic contextual understanding with physics-informed data consistency to deliver robust reconstructions across varied scanners, protocols, and patient presentations. Comprehensive evaluations demonstrate that CardioMM achieves state-of-the-art performance across internal centers and exhibits strong zero-shot generalization to unseen external settings. Importantly, CardioMM supports acceleration up to 24x, providing the first evidence that such extreme acquisition speed can preserve key cardiac phenotypes, quantitative myocardial biomarkers, and diagnostic image quality without compromising clinical integrity.

eess.IV↗

arXiv2025

Structural Reinforcement Learning for Heterogeneous Agent Macroeconomics

We present a new approach to formulating and solving heterogeneous agent models with aggregate risk. We replace the cross-sectional distribution with low-dimensional prices as state variables and let agents learn equilibrium price dynamics directly from simulated paths. To do so, we introduce a structural reinforcement learning (SRL) method which treats prices via simulation while exploiting agents' structural knowledge of their own individual dynamics. Our SRL method yields a general and highly efficient global solution method for heterogeneous agent models that sidesteps the Master equation and handles problems traditional methods struggle with, in particular nontrivial market-clearing conditions. We illustrate the approach in the Krusell-Smith model, the Huggett model with aggregate shocks, and a HANK model with a forward-looking Phillips curve, all of which we solve globally within minutes.

econ.TH↗

arXiv2025

Search for a solar-bound axion halo using the Global Network of Optical Magnetometers for Exotic physics searches

We report on a search for a gravitationally bound solar axion halo using data from the Global Network of Optical Magnetometers for Exotic physics searches (GNOME), a worldwide array of magnetically shielded atomic magnetometers with sensitivity to exotic spin couplings. Motivated by recent theoretical work suggesting that self-interacting ultralight axions can be captured by the Sun's gravitational field and thermalize into the ground state, we develop a signal model for the pseudo-magnetic fields generated by axion-proton gradient couplings in such a halo. The analysis focuses on the fifth GNOME Science Run (69 days, 12 stations), employing a cross-correlation pipeline with time-shifted daily modulation templates to search for the global, direction-dependent, monochromatic signal expected from a solar axion halo. No statistically significant candidate signals are observed. We set 95% confidence-level upper limits on the amplitude of the axion-induced pseudo-magnetic field over the frequency range $\approx 0.05-20$ Hz, translating to constraints on the linear and quadratic axion-proton couplings for halo densities predicted by gravitational capture models and for the maximum overdensities allowed by planetary ephemerides. In the quadratic coupling case, our limits surpass existing astrophysical bounds by over two orders of magnitude across much of the accessible parameter space.

physics.atom-ph↗

arXiv2025

High-current p-type transistors from precursor-engineered synthetic monolayer WSe$_2$

Monolayer tungsten diselenide (WSe$_2$) is a leading candidate for nanoscale complementary logic. However, high defect densities introduced during thin-film growth and device fabrication have limited p-type transistor performance. Here, we report a combined strategy of precursor-engineered chemical vapor deposition and damage-minimizing fabrication to overcome this limitation. By converting tungsten trioxide and residual oxyselenides into reactive suboxides before growth, and precisely regulating selenium delivery during deposition, we synthesize uniform, centimeter-scale monolayer WSe$_2$ films with charged defect densities as low as $5 \times 10^{9}$ cm$^{-2}$. Transistors fabricated from these films achieve record p-type on-state current up to $888 μ$A$\cdotμ$m$^{-1}$ at $V_{\mathrm{DS}}=-1$ V, matching leading n-type devices. This leap in material quality closes the p-type performance gap without exotic doping or contact materials, marking a critical step towards complementary two-dimensional semiconductor circuits.

cond-mat.mtrl-sci↗

arXiv2025

Task Adaptation from Skills: Information Geometry, Disentanglement, and New Objectives for Unsupervised Reinforcement Learning

Unsupervised reinforcement learning (URL) aims to learn general skills for unseen downstream tasks. Mutual Information Skill Learning (MISL) addresses URL by maximizing the mutual information between states and skills but lacks sufficient theoretical analysis, e.g., how well its learned skills can initialize a downstream task's policy. Our new theoretical analysis in this paper shows that the diversity and separability of learned skills are fundamentally critical to downstream task adaptation but MISL does not necessarily guarantee these properties. To complement MISL, we propose a novel disentanglement metric LSEPIN. Moreover, we build an information-geometric connection between LSEPIN and downstream task adaptation cost. For better geometric properties, we investigate a new strategy that replaces the KL divergence in information geometry with Wasserstein distance. We extend the geometric analysis to it, which leads to a novel skill-learning objective WSEP. It is theoretically justified to be helpful to downstream task adaptation and it is capable of discovering more initial policies for downstream tasks than MISL. We finally propose another Wasserstein distance-based algorithm PWSEP that can theoretically discover all optimal initial policies.

cs.LG↗

arXiv2025

A Lagrangian Approach to Optimal Randomization

We develop an efficient method for solving non-convex constrained optimization problems that are pervasive in economics. The optimal solution to these problems often involves randomization. We employ a Lagrangian framework and prove that the value of the saddle point characterizing the optimal random solution equals the value of the deterministic dual problem. Our algorithm solves this dual via subgradient descent and recovers the optimal random solution directly from deterministic optima computed along the iterations. For many non-convex economic problems, our method is orders of magnitude faster than linear programming, making previously intractable lottery problems feasible. As an application, we solve for optimal Mirrleesian income taxation with multi-dimensional types. We show that heterogeneity in productivity and Frisch elasticity can make randomization welfare-improving over the optimal deterministic schedule.

econ.TH↗

arXiv2024

Evaluating the Impact of Warning Modalities and False Alarms in Pedestrian Crossing Alert System

With the steadily increasing pedestrian fatalities, pedestrian safety is a growing concern, especially in urban environments. Advanced Driver Assistance Systems (ADAS) have been developed to mitigate road user risks by predicting potential pedestrian crossings and issuing timely driver alerts. However, there is limited understanding of how drivers respond to different modalities of alerts, particularly in the presence of false alarms. In this study, we utilized a full-scale driving simulator to compare the effectiveness of different alert modalities, audio-visual (AV), visual-tactile (VT), and audio-visual-tactile (AVT), in alerting drivers to various pedestrian jaywalking events. Our findings reveal that, compared to no alerts, multimodal alerts significantly increased the number of vehicles stopped for pedestrians and the distance to pedestrians when stopped. However, the false alarms negatively impacted driver trust, with some drivers exhibiting excessive caution, alert fatigue and anxiety, even including one instance where a driver fully stopped when no pedestrian was present.

cs.HC↗

arXiv2024

Reliable Heading Tracking for Pedestrian Road Crossing Prediction Using Commodity Devices

Pedestrian heading tracking enables applications in pedestrian navigation, traffic safety, and accessibility. Previous works, using inertial sensor fusion or machine learning, are limited in that they assume the phone is fixed in specific orientations, hindering their generalizability. We propose a new heading tracking algorithm, the Orientation-Heading Alignment (OHA), which leverages a key insight: people tend to carry smartphones in certain ways due to habits, such as swinging them while walking. For each smartphone attitude during this motion, OHA maps the smartphone orientation to the pedestrian heading and learns such mappings efficiently from coarse headings and smartphone orientations. To anchor our algorithm in a practical scenario, we apply OHA to a challenging task: predicting when pedestrians are about to cross the road to improve road user safety. In particular, using 755 hours of walking data collected since 2020 from 60 individuals, we develop a lightweight model that operates in real-time on commodity devices to predict road crossings. Our evaluation shows that OHA achieves 3.4 times smaller heading errors across nine scenarios than existing methods. Furthermore, OHA enables the early and accurate detection of pedestrian crossing behavior, issuing crossing alerts 0.35 seconds, on average, before pedestrians enter the road range.

eess.SP↗