arXiv subjects

Yukun Liu

Publications and source records attributed to Yukun Liu.

At least 19 recordsLinked to original sources

arXiv2026

The Cross-Section of Stock Returns and AI Exposure

We study 380 trillion tokens of realized AI consumption across more than four hundred LLMs. We build a high-frequency AI factor and show that a long-short strategy based on firms' AI exposure earns significantly positive returns. The average strategy return is larger based on intensive, frontier-oriented AI consumption but smaller based on casual or open-weight usage. Internationally, the return spread is significant in developed countries but insignificant in emerging markets. Examining occupational AI exposure, we find more positive exposure in occupations intensive in nonroutine interactive tasks and more negative exposure in those intensive in nonroutine analytical tasks.

cs.CY↗

arXiv2026

Massive Galaxy Halos Contain Less Inner Dark Matter Than Predicted

The mass profiles of galaxy halos encode how baryons reshape dark matter distribution, yet direct observational constraints across the full radial range remain scarce. Here we combine stellar kinematics from MaNGA, H I dynamical measurements from ALFALFA, and independently calibrated halo masses of SDSS groups to statistically reconstruct the mass distribution of central galaxies over nearly two orders of magnitude in radius. We demonstrate that H I data alone do not provide reliable total halo mass estimates, necessitating an independent group-based halo-mass scale. Compared to the IllustrisTNG and EAGLE simulations, the observational profiles of low-mass halos are broadly consistent; in contrast, massive observed halos exhibit systematically lower dynamical masses at the H I radius, lower inner dark-matter masses, and lower central dark-matter fractions (about 4$σ$ difference in units of population scatter) at fixed total halo mass. After subtracting baryonic contributions, the inferred dark-matter profiles remain broadly consistent with an NFW form, but with lower effective concentrations than predicted for massive halos. These results suggest that the inner dark-matter content of massive halos has been reduced more significantly than predicted by current hydrodynamical simulations, plausibly due to long-term baryonic halo heating in massive systems.

astro-ph.GA↗

arXiv2026

Reconstructing the Projected Dark Matter Field across 0.1-100 Mpc Scales from the SDSS Survey

Dark matter sets the gravitational environment in which galaxies form and evolve, but cannot be observed directly. We present a conditional diffusion model that reconstructs the projected dark matter density field from the galaxy stellar-mass density field for direct application to galaxy surveys. The model is trained on CAMELS and validated on the independent IllustrisTNG300-1 simulation. Halo masses inferred from the reconstructed projected-aperture measurements agree well with the corresponding true values, with a scatter below 0.2 dex. On 100 kpc scales, reconstructed surface densities show a typical scatter of ~0.3 dex in the regime most relevant for observations. We apply the model to SDSS galaxies with $M_\star\ge10^9\,M_\odot$ in a contiguous low-redshift region. Averaging over 100 stochastic realizations, we reconstruct and publicly release a projected dark matter field covering $90\times90\,(h^{-1}\mathrm{Mpc})^2$ with a pixel size of $0.097\,h^{-1}\mathrm{Mpc}$. This pixel area corresponds to the characteristic projected area of halos with masses of ~$10^{10.6}\,h^{-1}\,M_\odot$. The map reveals the multiscale projected cosmic web, including cluster-scale overdensities, filaments and voids. Projected-aperture masses are statistically consistent with SDSS group-catalog masses, while the derived halo mass function broadly matches mock-catalog expectations. The reconstructed projected potential places Coma in one of the deepest wells and near a convergence region of the inferred projected acceleration field, suggesting that the reconstruction retains both local overdensities and coherent large-scale projected gravitational structure. This work shows that diffusion-based dark matter reconstruction can be applied to real galaxy surveys, enabling halo-mass- and spatially resolved dark-matter-environment-based studies of galaxy evolution in SDSS and future wide-area surveys.

astro-ph.GA↗

arXiv2026

No Detectable One-halo Galactic Conformity Signal with Halo-mass Estimates Consistent with Weak-lensing Constraints

One-halo galactic conformity is the tendency for satellites in halos with quenched centrals to have lower star-formation activity than those in halos with star-forming centrals at fixed halo mass. It is an important probe of the galaxy--halo connection and halo-wide quenching processes that may couple central and satellite evolution. However, its existence remains controversial, because conformity must be measured at fixed halo mass, while halo masses are difficult to estimate accurately. In this Letter, we measure one-halo conformity in SDSS using five stellar-mass-complete samples and three halo-mass estimates: an ML estimate whose star-forming and quenched stellar mass--halo mass relations (SHMRs) agree with independent weak-lensing constraints, and two conventional abundance-matching (AM) estimates. We quantify conformity as the difference in median $\log({\rm sSFR})$ between satellites of star-forming and quenched centrals, using both satellite-level and halo-level statistics. The two AM estimates produce strong positive conformity signals, consistent with previous AM-based measurements, but these signals are not reproduced with the ML halo masses. For the halo-level statistic, the representative AM-based signals are $+0.38\pm0.04$ dex and $+0.23\pm0.04$ dex for the luminosity-ranking and mass-ranking AM halo masses, detected relative to no conformity at about $10σ$ and $6σ$, respectively. In contrast, the ML result is consistent with no conformity, $+0.00\pm0.03$ dex; the satellite-level statistic gives a similar result. Thus, with halo-mass estimates consistent with weak-lensing constraints, we find no detectable one-halo conformity signal in the present SDSS sample, suggesting that the strong AM-based signal is largely driven by halo-mass estimation biases.

astro-ph.GA↗

arXiv2026

Who Remains, What Changes: Identity Anchored Composed Gait Retrieval

Gait recognition has achieved remarkable progress, yet existing methods remain confined to rigid visual matching and often overlook the potential of natural language instructions for interactive retrieval. In this paper, we introduce Composed Gait Retrieval (CoGR), a novel task that retrieves a target gait sequence based on a reference sequence and a natural language modification query. To address the absence of existing datasets for this task, we design an automated annotation pipeline powered by large vision-language models (VLMs) to construct the first gait-language datasets: Language-Augmented CCPG and Language-Augmented CASIA-B. Building on this, we propose ComposeGait, an identity-anchored composition framework designed to prevent the identity drift that arises when generic composed retrieval follows the instruction but returns the wrong person. Its Part-aware Identity Adapter (PIA) aggregates multi-frame, part-aware identity evidence into a sample-specific ID token. We inject the ID tokens into both branches of a shared Q-Former to preserve identity, while excluding the ID-token outputs from the final retrieval embeddings. Joint identity and task-adapted composed-retrieval objectives optimize this space end to end. We evaluate ComposeGait on both benchmarks and show that it achieves the best R@1 among the compared methods, reaching 72.38% on Language-Augmented CCPG and 83.61% on Language-Augmented CASIA-B. These results establish ComposeGait as a strong baseline for CoGR. The datasets and code will be made publicly available.

cs.CV↗

arXiv2026

Physics-Informed Neural Networks for Sparse Strain-Field Reconstruction in 4D-STEM

Quantitative strain mapping using four-dimensional scanning transmission electron microscopy (4D-STEM) typically requires densely sampled scans that can damage beam-sensitive specimens. We develop a physics-informed neural network (PINN) for sparse 4D-STEM strain reconstruction that embeds elastic equilibrium and Saint-Venant compatibility in the training loss through automatic differentiation. The architecture combines a coordinate-based implicit representation, sine activations with stable second derivatives, frozen residual-scale normalization, an exponential physics-weight ramp, and residual-based adaptive collocation. We apply a sine-activated residual network to an experimental $180\times400$-pixel strain map of domain-structured PbGeSnSe$_{1.5}$Te$_{1.5}$. Across $1$-$75%$ sampling ($720$-$54{,}000$ probe positions), $R^2$ for $\varepsilon_{xx}$ reaches $0.80$ at $10%$ sampling and saturates near $0.86$ by $25%$; the chevron strain-band morphology is recovered from $10%$ of probe positions. At $10%$ sampling, the PINN reduces mean absolute error by approximately $26%$ relative to compressed sensing and $22%$ relative to Gaussian-process regression. An ablation against an equal-capacity data-only SIREN shows that the PDE prior improves accuracy at extreme sparsity and consistently improves physical self-consistency, but biases the reconstruction when data are abundant. Monte Carlo dropout and mean-field variational inference provide per-pixel epistemic uncertainty maps correlated with reconstruction error. With an appropriate constitutive model, the framework is adaptable to strain mapping across diverse material systems.

cond-mat.mtrl-sci↗

arXiv2026

Trading Frictions in Dynamic Cap-and-Trade Markets

We develop a dynamic stochastic model of markets with an externality and multiple trading frictions, and cap-and-trade as the leading application. Slow participation, limited intermediation, and heterogeneous information interact in equilibrium: agents choose costly market access, access determines residual compliance demand, intermediary constraints translate residual demand into a surrender-month premium, and the premium feeds back into access incentives. These interactions shape how effectively the market corrects the externality. We characterize access choices in closed form, prove that the equilibrium premium is unique, and show that endogenous access dampens the response to each friction in isolation, while the interaction of multiple frictions is non-additive and can amplify the price response. We quantify the model using 2.7 million EU ETS registry transactions and compliance records from 2005-2021. About 40% of operators do not trade annually, purchases concentrate in April when returns are systematically high, and operator flow predicts future returns.

econ.TH↗

arXiv2026

Higher-Order Asset Pricing Factors via Forward Selection Fama-MacBeth Regression

We show that the higher-order terms and interactions of the common sparse linear factors are significantly priced in the cross-section of equity returns. A higher-order model with only a small number of selected higher-order terms from six widely used factors outperforms traditional benchmarks both in-sample and out-of-sample. It also substantially reduces the alphas of the extensive factor zoo, suggesting that the pricing power of many zoo factors is attributable to their exposure to higher-order terms of common linear factors. We identify and rank the most relevant higher-order terms by developing a forward selection Fama-MacBeth procedure.

econ.EM↗

arXiv2026

Cryptocurrency as an Investable Asset Class: Coming of Age

We organize existing empirical regularities of cryptocurrencies into seven stylized facts and analyze cryptocurrencies through the lens of empirical asset pricing. We find important similarities with traditional markets--risk-adjusted performance so far is broadly comparable, and the cross-section of returns can be summarized by a small set of factors. However, cryptocurrency also has its own distinct character: jumps are frequent and large, and blockchain information helps drive prices. This common set of stylized facts provides evidence that cryptocurrency is emerging as an investable asset class. Additionally, we discuss potential data quality issues and possible changes in future regulations and the cryptocurrency environment.

q-fin.GN↗

arXiv2026

SecCodeBench-V2 Technical Report

We introduce SecCodeBench-V2, a publicly released benchmark for evaluating Large Language Model (LLM) copilots' capabilities of generating secure code. SecCodeBench-V2 comprises 98 generation and fix scenarios derived from Alibaba Group's industrial productions, where the underlying security issues span 22 common CWE (Common Weakness Enumeration) categories across five programming languages: Java, C, Python, Go, and JavaScript. SecCodeBench-V2 adopts a function-level task formulation: each scenario provides a complete project scaffold and requires the model to implement or patch a designated target function under fixed interfaces and dependencies. For each scenario, SecCodeBench-V2 provides executable proof-of-concept (PoC) test cases for both functional validation and security verification. All test cases are authored and double-reviewed by security experts, ensuring high fidelity, broad coverage, and reliable ground truth. Beyond the benchmark itself, we build a unified evaluation pipeline that assesses models primarily via dynamic execution. For most scenarios, we compile and run model-generated artifacts in isolated environments and execute PoC test cases to validate both functional correctness and security properties. For scenarios where security issues cannot be adjudicated with deterministic test cases, we additionally employ an LLM-as-a-judge oracle. To summarize performance across heterogeneous scenarios and difficulty levels, we design a Pass@K-based scoring protocol with principled aggregation over scenarios and severity, enabling holistic and comparable evaluation across models. Overall, SecCodeBench-V2 provides a rigorous and reproducible foundation for assessing the security posture of AI coding assistants, with results and artifacts released at https://alibaba.github.io/sec-code-bench. The benchmark is publicly available at https://github.com/alibaba/sec-code-bench.

cs.CR↗

arXiv2026

Beyond Aggregation: Guiding Clients in Heterogeneous Federated Learning

Federated learning (FL) is increasingly adopted in domains like healthcare, where data privacy is paramount. A fundamental challenge in these systems is statistical heterogeneity-the fact that data distributions vary significantly across clients (e.g., different hospitals may treat distinct patient demographics). While current FL algorithms focus on aggregating model updates from these heterogeneous clients, the potential of the central server remains under-explored. This paper is motivated by a healthcare scenario: could a central server not only coordinate model training but also guide a new patient to the hospital best equipped for their specific condition? We generalize this idea to propose a novel paradigm for FL systems where the server actively guides the allocation of new tasks or queries to the most appropriate client. To enable this, we introduce a density ratio model and empirical likelihood-based framework that simultaneously addresses two goals: (1) learning effective local models on each client, and (2) finding the best matching client for a new query. Empirical results demonstrate the framework's effectiveness on benchmark datasets, showing improvements in both model accuracy and the precision of client guidance compared to standard FL approaches. This work opens a new direction for building more intelligent and resource-efficient FL systems that leverage heterogeneity as a feature, not just a bug. Code is available at https://github.com/zijianwang0510/FedDRM.git.

cs.LG↗

arXiv2026

Minimum Wasserstein distance estimator under covariate shift: closed-form, super-efficiency and irregularity

Covariate shift arises when covariate distributions differ between source and target populations while the conditional distribution of the response remains invariant, and it underlies problems in missing data and causal inference. We propose a minimum Wasserstein distance estimation framework for inference under covariate shift that avoids explicit modeling of outcome regressions or importance weights. The resulting W-estimator admits a closed-form expression and is numerically equivalent to the classical 1-nearest neighbor estimator, yielding a new optimal transport interpretation of nearest neighbor methods. We establish root-$n$ asymptotic normality and show that the estimator is not asymptotically linear, leading to super-efficiency relative to the semiparametric efficient estimator under covariate shift in certain regimes, and uniformly in missing data problems. Numerical simulations, along with an analysis of a rainfall dataset, underscore the exceptional performance of our W-estimator.

stat.ME↗

arXiv2025

Breaking Data Silos: Towards Open and Scalable Mobility Foundation Models via Generative Continual Learning

Human mobility is a fundamental pillar of urban science and sustainability, providing critical insights into energy consumption, carbon emissions, and public health. However, the discovery of universal mobility laws is currently hindered by the ``data silo'' problem, where institutional boundaries and privacy regulations fragment the necessary large-scale datasets. In this paper, we propose MoveGCL, a transformative framework that facilitates collaborative and decentralized mobility science via generative continual learning. MoveGCL enables a distributed ecosystem of data holders to jointly evolve a foundation model without compromising individual privacy. The core of MoveGCL lies in its ability to replay synthetic trajectories derived from a generative teacher and utilize a mobility-pattern-aware Mixture-of-Experts (MoE) architecture. This allows the model to encapsulate the unique characteristics of diverse urban structures while mitigating the risk of knowledge erosion (catastrophic forgetting). With a specialized layer-wise progressive adaptation strategy, MoveGCL ensures stable convergence during the continuous integration of new urban domains. Our experiments on six global urban datasets demonstrate that MoveGCL achieves performance parity with joint training, a previously unattainable feat under siloed conditions. This work provides a scalable, privacy-preserving pathway toward Open Mobility Science, empowering researchers to address global sustainability challenges through cross-institutional AI collaboration. To facilitate reproducibility and future research, we have released the code and models at \color{blue}{https://github.com/tsinghua-fib-lab/MoveGCL}.

cs.LG↗

arXiv2025

M3-TTS: Multi-modal DiT Alignment & Mel-latent for Zero-shot High-fidelity Speech Synthesis

Non-autoregressive (NAR) text-to-speech synthesis relies on length alignment between text sequences and audio representations, constraining naturalness and expressiveness. Existing methods depend on duration modeling or pseudo-alignment strategies that severely limit naturalness and computational efficiency. We propose M3-TTS, a concise and efficient NAR TTS paradigm based on multi-modal diffusion transformer (MM-DiT) architecture. M3-TTS employs joint diffusion transformer layers for cross-modal alignment, achieving stable monotonic alignment between variable-length text-speech sequences without pseudo-alignment requirements. Single diffusion transformer layers further enhance acoustic detail modeling. The framework integrates a mel-vae codec that provides 3* training acceleration. Experimental results on Seed-TTS and AISHELL-3 benchmarks demonstrate that M3-TTS achieves state-of-the-art NAR performance with the lowest word error rates (1.36\% English, 1.31\% Chinese) while maintaining competitive naturalness scores. Code and demos will be available at https://wwwwxp.github.io/M3-TTS.

cs.SD↗

arXiv2025

UniMove: A Unified Model for Multi-city Human Mobility Prediction

Human mobility prediction is vital for urban planning, transportation optimization, and personalized services. However, the inherent randomness, non-uniform time intervals, and complex patterns of human mobility, compounded by the heterogeneity introduced by varying city structures, infrastructure, and population densities, present significant challenges in modeling. Existing solutions often require training separate models for each city due to distinct spatial representations and geographic coverage. In this paper, we propose UniMove, a unified model for multi-city human mobility prediction, addressing two challenges: (1) constructing universal spatial representations for effective token sharing across cities, and (2) modeling heterogeneous mobility patterns from varying city characteristics. We propose a trajectory-location dual-tower architecture, with a location tower for universal spatial encoding and a trajectory tower for sequential mobility modeling. We also design MoE Transformer blocks to adaptively select experts to handle diverse movement patterns. Extensive experiments across multiple datasets from diverse cities demonstrate that UniMove truly embodies the essence of a unified model. By enabling joint training on multi-city data with mutual data enhancement, it significantly improves mobility prediction accuracy by over 10.2\%. UniMove represents a key advancement toward realizing a true foundational model with a unified architecture for human mobility. We release the implementation at https://github.com/tsinghua-fib-lab/UniMove/.

cs.LG↗

arXiv2025

Conformal Prediction Beyond the Horizon: Distribution-Free Inference for Policy Evaluation

Reliable uncertainty quantification is crucial for reinforcement learning (RL) in high-stakes settings. We propose a unified conformal prediction framework for infinite-horizon policy evaluation that constructs distribution-free prediction intervals {for returns} in both on-policy and off-policy settings. Our method integrates distributional RL with conformal calibration, addressing challenges such as unobserved returns, temporal dependencies, and distributional shifts. We propose a modular pseudo-return construction based on truncated rollouts and a time-aware calibration strategy using experience replay and weighted subsampling. These innovations mitigate model bias and restore approximate exchangeability, enabling uncertainty quantification even under policy shifts. Our theoretical analysis provides coverage guarantees that account for model misspecification and importance weight estimation. Empirical results, including experiments in synthetic and benchmark environments like Mountain Car, show that our method significantly improves coverage and reliability over standard distributional RL baselines.

stat.ML↗

arXiv2025

Semiparametric Learning from Open-Set Label Shift Data

We study the open-set label shift problem, where the test data may include a novel class absent from training. This setting is challenging because both the class proportions and the distribution of the novel class are not identifiable without extra assumptions. Existing approaches often rely on restrictive separability conditions, prior knowledge, or computationally infeasible procedures, and some may lack theoretical guarantees. We propose a semiparametric density ratio model framework that ensures identifiability while allowing overlap between novel and known classes. Within this framework, we develop maximum empirical likelihood estimators and confidence intervals for class proportions, establish their asymptotic validity, and design a stable Expectation-Maximization algorithm for computation. We further construct an approximately optimal classifier based on posterior probabilities with theoretical guarantees. Simulations and a real data application confirm that our methods improve both estimation accuracy and classification performance compared with existing approaches.

stat.ME↗

arXiv2025

Retrospective score tests versus prospective score tests for genetic association with case-control data

Since the seminal work by Prentice and Pyke (1979), the prospective logistic likelihood has become the standard method of analysis for retrospectively collected case-control data, in particular for testing the association between a single genetic marker and a disease outcome in genetic case-control studies. When studying multiple genetic markers with relatively small effects, especially those with rare variants, various aggregated approaches based on the same prospective likelihood have been developed to integrate subtle association evidence among all considered markers. In this paper we show that using the score statistic derived from a prospective likelihood is not optimal in the analysis of retrospectively sampled genetic data. We develop the locally most powerful genetic aggregation test derived through the retrospective likelihood under a random effect model assumption. In contrast to the fact that the disease prevalence information cannot be used to improve the efficiency for the estimation of odds ratio parameters in logistic regression models, we show that it can be utilized to enhance the testing power in genetic association studies. Extensive simulations demonstrate the advantages of the proposed method over the existing ones. One real genome-wide association study is analyzed for illustration.

stat.ME↗