Search arXiv⌕ Search

arXiv subjects

Jia Liu

Publications and source records attributed to Jia Liu.

At least 37 records · Page 2Linked to original sources
arXiv2026

Human-Level Accuracy, Non-Human Strategies: Revealing Model-Human Divergence in Video Physical Reasoning

Video foundation models now reach human-level accuracy on physical-reasoning benchmarks, yet such tasks require predicting unobserved physical outcomes. Do these models perform human-like forward simulation, or do they exploit statistical regularities in visible scenes? Accuracy alone cannot distinguish these strategies. We introduce a distributional evaluation framework that treats model seeds and human raters as populations, enabling comparison of consensus, uncertainty, and strategy. On the Physion benchmark, we evaluate three ViT-L architectures (V-JEPA2, VideoMAEv2, DINOv2). V-JEPA2 narrows the accuracy gap to ~1 percentage point (73.2% vs. 74.2%), yet model-human disagreement reaches 26.4%, far exceeding human-human disagreement (4.8%), with substantially lower agreement (kappa ~ 0.48 vs. 0.91). The divergence follows forward-simulation demands: models outperform humans on geometric reasoning (linking, +11.8 pp) but underperform on gravitational dynamics (rolling, -11.8 pp) and causal chains (dominoes, -10.5 pp). Strategy fingerprinting confirms all three architectures share non-human strategies while none aligns with humans. Attribution analysis suggests that unobservable outcome features, rather than visible scene properties, predict this divergence, consistent with models relying more on scene-level statistical regularities than on explicit forward simulation, a systematic divergence that accuracy alone cannot reveal. Code is available at https://github.com/fanhong-li/model-human-divergence.

cs.CV↗
arXiv2026

Future of Artificial Intelligence for Science in Japan 2024 Community Report

This white paper summarizes scientific challenges and AI/ML research opportunities identified through the FAIRS Japan 2024 unconference process. The discussion focuses on three major physics domains: accelerator physics, cosmology and astrophysics, and neutrino physics. Although each domain has distinct scientific goals and experimental constraints, several common technical themes emerge: high-dimensional reconstruction, fast and accurate simulation, uncertainty propagation, simulation-to-data mismatch, anomaly detection, real-time decision-making, and shared infrastructure.

hep-ph↗
arXiv2026

Cosmological Constrained Axion-Portal Inelastic Dark Matter for the LZ Event

The recent high-recoil candidate event reported by LUX-ZEPLIN (LZ) motivates dark-matter scenarios with nonstandard kinematics and momentum-dependent interactions. We study a two-state inelastic dark-matter model in which $χ_1$ and $χ_2$ couple off-diagonally to an axion-like particle (ALP) that also couples to gluons and photons. Unlike treatments that assume only one dark-matter state is present today, we track the cosmological evolution of both states. The excited state $χ_2$ is sufficiently long-lived to survive to the present, with its relic fraction determined by the dark-sector conversion process $χ_2χ_2\leftrightarrowχ_1χ_1$. We find that this fraction depends strongly on the Lorentz structure of the DM--ALP interaction: scalar transition couplings efficiently deplete $χ_2$, favoring endothermic $χ_1 N\toχ_2 N$ scattering, whereas pseudoscalar transition couplings preserve $f_2\simeq1/2$, yielding recoil spectra dominated by exothermic $χ_2 N\toχ_1 N$ down-scattering. Benchmark spectra in all four scenarios can peak near the observed recoil energy of $250\,\mathrm{keV}$. Our results demonstrate that the dominant direction of inelastic scattering in direct detection can be dynamically selected by the early-Universe evolution of the dark sector. With additional data, annual modulation measurements could distinguish these scenarios. We further show that the ALP portal can be directly probed at colliders.

hep-ph↗
arXiv2026

Hyperparameter Scaling Laws Across MoE Sparsity

Mixture-of-Experts (MoE) models expand model capacity without a proportional increase in training compute, but increasing sparsity makes reliable hyperparameter transfer challenging. In this work, we show that conventional hyperparameter scaling laws are insufficient for ultra-sparse MoEs: the optimal learning rate and batch size vary with activation ratio, and these shifts cannot be explained by either total or activated parameter count alone. To characterize this dependence, we conduct 1,800 pre-training runs spanning six activated-parameter scales and models with up to 6B total non-embedding parameters, processing approximately 20 trillion tokens at a cost of 200,000 equivalent H800 GPU-hours. Our results reconcile conflicting findings in prior work by revealing two scaling regimes. At fixed sparsity, the optimal batch size follows a power-law relationship with training tokens $D$, whereas the optimal learning rate scales with training compute $C$ and remains robust to the allocation between model size and data. Across sparsity levels, the activation ratio $A$ enters both relationships as an additional multiplicative power-law factor. These observations lead to unified hyperparameter scaling laws that transfer across MoE sparsity levels. Large-scale evaluation shows that the scaling form outperforms alternative functional forms. On a held-out ultra-sparse MoE with 12B total parameters and only 1/64 of its experts activated, the predicted hyperparameters remain close to the observed optima, supporting joint extrapolation across model scale and sparsity. Further experiments demonstrate transfer across expert granularities and isolate the effect of activation ratio from that of total expert count.

cs.LG↗
arXiv2026

SuperValid: Capability-Aligned OOD Validation for Generalizable Downstream Scaling

Scaling laws guide large language model training by relating compute to cross-entropy loss, and recent work further extends them to predict downstream benchmark performance. However, prior approaches face generalization limitations from two aspects: focusing on benchmark-level performance introduces scenario-specific artifacts, while relying on IID validation loss fails to track capability improvements when training distributions vary. In this work, we argue that downstream scaling should be studied at the capability level, which captures shared skill factors across related tasks while abstracting away benchmark-specific noise. We propose SuperValid, a framework that synthesizes OOD (out-of-distribution), capability-aligned validation data by distilling core concepts from benchmarks within a capability domain and expanding them into diverse, knowledge-rich texts. Extensive experiments spanning 16 benchmarks grouped into 6 capability domains show that SuperValid loss exhibits strong and stable correlation with downstream performance across models of different architectures, scales, and training data distributions. As a training-free metric computable during training without benchmark evaluation, SuperValid enables effective model selection, early stopping, and scaling decisions.

cs.CL↗
arXiv2026

Microscopic Calculation of Electric Quadrupole Effective Charges in Exotic Nuclei

Electric quadrupole ($E2$) effective charges are evaluated based on the self-consistent relativistic Hartree-Fock single-particle states, with core-polarization corrections resummed to all orders using the Tamm-Dancoff approximation (TDA). Configuration-interaction relativistic Hartree-Fock (CI-RHF) calculations employing the TDA effective charges well reproduce the $B(E2)$ strength for neon isotopes from stability to the neutron drip line. We find that polarization charges associated with continuum states are significantly quenched due to their extended density distributions and weak coupling to the core, underscoring the critical role of continuum effects in $E2$ transition evaluations for exotic nuclei. Moreover, the CI-RHF model predicts a suppressed $B(E2; 2^+_2 \to 0^+_2)$ in $^{30}$Ne, together with strong in-band $B(E2)$ strengths of the yrast band, suggesting the coexistence of a nearly spherical excited $0^+_2$ state and a deformed ground state within the $N=20$ "island of inversion".

nucl-th↗
arXiv2026

B(E2) Serves as a Robust Signature of N = 32,34 Shell Evolution

Electric quadrupole transition probabilities $B(E2)$ serve as key probe of nuclear shell evolution, yet anomalous $B(E2)$ values in exotic nuclei complicate the identification of new magic numbers. In this letter, employing the configuration-interaction relativistic Hartree-Fock model, we demonstrate that effective charges are sensitive to orbital radii, and this orbital dependence is significantly amplified by the halo structure of valence nucleons. This mechanism is critical for reliably describing $E2$ transitions and understanding the unusual behavior of $B(E2)$ in exotic nuclei. Our calculations predict reduced $B(E2; 2^+_1 \rightarrow 0^+_1)$ values in $^{52,54}\text{Ca}$, signaling the emergence of subshell closures at $N=32$ and 34. Furthermore, the suppressed $B(E2; 7/2^{-}_{1} \rightarrow 11/2^{-}_{1})$ transition in $^{53}\text{Sc}$ underscores the robustness of the $N=32$ new magic number, whereas the enhanced transition strength in $^{55}\text{Sc}$ indicates the rapid erosion of the $N=34$ shell gap with the occupancy of the proton orbital $\pi1f_{7/2}$.

nucl-th↗
arXiv2026

Latency-Response Theory Model: Evaluating Large Language Models via Response Accuracy and Chain-of-Thought Length

The proliferation of Large Language Models (LLMs) necessitates valid evaluation methods to provide guidance for both downstream applications and actionable future improvements. The Item Response Theory (IRT) model with Computerized Adaptive Testing has recently emerged as a promising framework for evaluating LLMs via their response accuracy. Beyond simple response accuracy, LLMs' chain of thought (CoT) lengths serve as a vital indicator of their reasoning ability. To leverage the CoT length information to assist LLM evaluation, we propose the \textbf{La}tency-\textbf{R}esponse \textbf{T}heory (LaRT) model, which jointly models both the response accuracy and CoT length by introducing a key correlation parameter between the latent ability and the latent speed. We derive an efficient stochastic approximation Expectation-Maximization algorithm for parameter estimation. We establish rigorous identifiability results for the latent ability and latent speed parameters to ensure the statistical validity of their estimation. Through both theoretical asymptotic analyses and simulation studies, we demonstrate LaRT's advantages over IRT in terms of superior estimation accuracy and shorter confidence intervals for latent trait estimation. To evaluate LaRT in real data, we collect responses from diverse LLMs on popular benchmark datasets. We find that LaRT yields different LLM rankings than IRT and outperforms IRT across multiple key evaluation metrics including predictive power, item efficiency, ranking validity, and LLM evaluation efficiency. Code and data are available at https://github.com/Toby-X/Latency-Response-Theory-Model

stat.ME↗
arXiv2026

SORT: A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders

While Transformers have achieved remarkable success in LLMs through superior scalability, their application in industrial-scale ranking models remains nascent, hindered by the challenges of high feature sparsity and low label density. In this paper, we propose SORT (Systematically Optimized Ranking Transformer), a scalable model designed to bridge the gap between Transformers and industrial-scale ranking models. We address the high feature sparsity and low label density challenges through a series of optimizations, including request-centric sample organization, local attention, query pruning, and generative pre-training. Furthermore, we introduce a suite of refinements to the tokenization, multi-head attention (MHA), and feed-forward network (FFN) modules, which collectively stabilize the training process and enlarge the model capacity. To maximize hardware efficiency, we optimize our training system to elevate the model FLOPs utilization (MFU) to 45%. Extensive experiments demonstrate that SORT outperforms strong baselines and exhibits excellent scalability across data size, model size, and sequence length, while remaining flexible at integrating diverse features. Finally, online A/B testing in large-scale e-commerce scenarios confirms that SORT achieves significant gains in key business metrics, including orders (+7.47%), buyers (+6.67%), and GMV (+8.65%), while simultaneously cutting latency by 62% and boosting throughput nearly sevenfold (+589%). SORT has been fully deployed in production, serving all users on AliExpress.

cs.IR↗
arXiv2026

A Synthetic Iterative Scheme for Non-Gray Phonon Boltzmann Transport Equation with Dual Relaxation Times

Solving the non-gray Callaway phonon Boltzmann transport equation allows a dual-relaxation-time approximation of separate normal and resistive scatterings and resolving the mode-dependent spectrum. The conventional iterative scheme (CIS) for deterministic solutions avoids a monolithic phase-space inversion, but its collision-source iteration can become prohibitively slow at a large characteristic length of a material. Existing synthetic acceleration schemes address either non-gray single-relaxation models or gray dual-relaxation models, leaving mode-resolved dual-relaxation transport without a dedicated acceleration framework. We develop a general synthetic iterative scheme (GSIS) for the stationary, linearized, non-gray Callaway equation, where synthetic approximations for the normal-process pseudo-temperature and phonon drift velocity are provided by exact energy and quasi-momentum balance laws closed with first-order Chapman-Enskog constitutive relations and non-equilibrium terms evaluated from the kinetic solution. The resistive-process pseudo-temperature is retrieved from the two quantities. Each iteration couples an upwind nodal discontinuous Galerkin kinetic sweep and a hybridizable discontinuous Galerkin solution of the synthetic equations to achieve high-order spatial discretization. A branch- and frequency-resolved Fourier analysis identifies the deterioration of CIS and shows that the GSIS contraction factor remains bounded away from unity for the considered graphene material. Asymptotic analysis indicates that GSIS reduces to a consistent discretization of a Guyer-Krumhansl-like equation and Fourier's law of heat conduction in the hydrodynamic and diffusive limits, respectively.

physics.comp-ph↗
arXiv2026

A Scalable Lift-and-Project Differentiable Approach For the Maximum Cut Problem

We propose a scalable framework for solving the Maximum Cut (MaxCut) problem in large graphs using projected gradient ascent on quadratic objectives. Our approach is differentiable and leverages GPUs for gradient-based optimization. It is not a machine learning method and does not require training data. Starting from a continuous relaxation of the classical quadratic binary formulation, we present a parallelized strategy that explores multiple initialization vectors in batch. We analyze the relaxed objective, showing it is convex and has fixed-points corresponding to local optima, particularly at boundary points, highlighting a key challenge in non-convex optimization. To improve exploration, we introduce a lifted quadratic formulation that over-parameterizes the solution space. We also provide a theoretical characterization of these lifted fixed-points. Finally, we propose DECO, a dimension-alternating algorithm that switches between the unlifted and lifted formulations, combined with importance-based degree initialization and a population-based evolutionary hyper-parameter search. Experiments on diverse graph families show that our methods attain comparable or superior performance relative to recent neural networks and GPU-accelerated sampling approaches.

cs.DM↗
arXiv2026

An Axial $U_A(1)_{L_μ-L_τ}$: UV Completion and Experimental Searches

We propose an anomaly-free and renormalizable axial $U_A(1)_{L_μ-L_τ}$ model and study its experimental signatures for $A'$ masses from the MeV scale to the TeV scale. The opposite charges of the left- and right-handed charged leptons forbid the usual muon and tau Yukawa interactions. Their masses are instead generated by a singlet scalar and heavy vector-like leptons through a universal-seesaw mechanism. We focus on heavy vector-like leptons, small light--heavy mixing, and $m_s\gtrsim10~\mathrm{GeV}$. In this limit, the observables considered here depend mainly on $(m_{A'},g_X)$, while the other model parameters are restricted by mixing and perturbativity. We confront this benchmark with current experimental searches. For neutrino trident production, our finite-$m_μ$ calculation shows that $A'$ modifies the axial weak coefficient, rather than the vector coefficient relevant to the usual $L_μ-L_τ$ model. The longitudinal mode enhances muon bremsstrahlung and gives a negative contribution to $(g-2)_μ$; the latter dominates over the scalar contribution in our benchmark. Combining these results with invisible meson decays and four-muon resonance searches, we summarize the phenomenological constraints in the $(m_{A'},g_X)$ plane. For $m_{A'}\gg m_μ$, vector and axial final-state-radiation rates become nearly identical, so the corresponding collider limits can be obtained by rate matching. At a future muon collider, the total rate alone does not fully resolve the interaction structure, whereas angular distributions, especially the forward--backward asymmetry in $μ^+μ^-\toτ^+τ^-$, retain direct sensitivity to chirality.

hep-ph↗
arXiv2026

Topological phase rectification via Aharonov-Bohm interference in a Majorana--quantum-dot interferometer

We propose and theoretically investigate a topological superconducting rectifier based on a quantum-dot--Majorana interferometer. The Aharonov-Bohm phase, controlled by a magnetic flux threading the interferometer loop, tunes the quantum interference between a trivial $2π$-periodic quantum-dot channel and a topological $4π$-periodic Majorana channel. At non-integer flux, this interference generates a persistent current background $I_{\rm off}$ that shifts the current-phase relation into a unipolar regime, in which the supercurrent flows strictly in one direction. We introduce a signed unipolarity factor $η_u$, with $|η_u|>0.5$ defining the unipolar regime, and establish its quantitative relationship to the conventional diode efficiency $η$. The unipolarity proves robust against variations of the quantum-dot level, spin polarization, and Majorana hybridization, is enhanced by stronger Majorana coupling and Rashba spin-orbit interaction, and persists at realistic temperatures and under quasiparticle poisoning. We further propose a topological diode figure of merit $\mathcal{Z}_{\rm TD}$, defined from the Fourier spectrum of $η_u$, whose nonzero value provides a model-independent signature of the $4π$-periodic Majorana channel and distinguishes topological from trivial rectification mechanisms. Our findings establish the quantum-dot--Majorana interferometer as a promising route toward high-performance topological superconducting diodes with clear experimental signatures accessible via standard dc transport measurements.

cond-mat.mes-hall↗
arXiv2026

Flavor--Kinetic Entanglement Production from Decay and Scattering at Finite Density

We extend the scattering-entanglement dictionary to finite-density environments by investigating the flavor--kinetic bipartition of the Hilbert space. We show that tracing over kinematic degrees of freedom maps the total branch-changing transition probability directly onto the leading flavor--kinetic linear entanglement entropy. At finite density, the vacuum branch-changing probability is replaced by an occupation-weighted collision probability, built from the same directed reaction-density kernel that enters the integrated Boltzmann equation. The resulting observable is the bath-averaged flavor--kinetic entanglement entropy of a pair sampled from the medium. As a proof of principle, this framework is applied to an $O(N)$ singlet-scalar extended model to probe thermal phase transitions. In the examples studied, the resulting entanglement entropy serves as a collision-based phase-transition-type diagnostic, exhibiting a finite discontinuity across a first-order phase transition and a nonanalytic temperature derivative for continuous transitions. These examples suggest a novel way to characterize thermal phase structures, distinct from traditional thermodynamic order parameters.

hep-ph↗
arXiv2026

AIvilization v0: Toward Large-Scale Artificial Social Simulation with a Unified Agent Architecture and Adaptive Agent Profiles

AIvilization v0 is a publicly deployed large-scale artificial society that couples a resource-constrained sandbox with a unified LLM-agent architecture, aiming to sustain long-horizon autonomy while remaining executable under a rapidly changing environment. To mitigate the tension between goal stability and reactive correctness, keeping long-horizon objectives on course while each action remains valid in a fast-changing shared world, we introduce (i) a hierarchical branch-thinking planner that decomposes life goals into parallel objective branches and uses simulation-guided validation plus tiered re-planning to ensure feasibility; (ii) an adaptive agent profile with dual-process memory that separates short-term execution traces from long-term semantic consolidation, enabling persistent yet evolving identity; and (iii) a human-in-the-loop steering interface that injects long-horizon objectives and short commands at appropriate abstraction levels, with effects propagated through memory instead of brittle prompt overrides. The environment integrates physiological survival costs, non-substitutable multi-tier production, an AMM-based price mechanism, and a gated education-occupation system. In a large-scale public deployment with tens of thousands of agents, high-frequency transactions from the platform's mature phase reveal stable markets that reproduce key stylized facts of real economies and structured wealth stratification driven by education and access constraints. At the agent level, portraits evolve coherently over long horizons, and human steering is associated with measurably larger short-horizon profile updates. Controlled ablation experiments complement the deployment evidence, showing that our agent architecture is robust in multi-objective, long-horizon settings.

cs.MA↗
arXiv2026

Generation of high-fluence and high-intensity hard x-ray attosecond pulses at European XFEL

By combining hard x-ray attosecond pulses from the European XFEL with total-reflection focusing x-ray optics, we generated nanofocused hard x-ray attosecond pulses with intensities and fluences comparable to the highest values attained in the hard x-ray regime. A peak intensity on the order of 10$^{20}$ W/cm$^2$ is confirmed through the observation of saturation in amplified spontaneous emission from copper atoms. These x-ray pulses enable new scientific opportunities, including the exploration of higher-order nonlinear light--matter interactions, damage-free structure determination, and coherent control of atoms and molecules.

physics.optics↗
arXiv2026

Replicating weak-lensing summary-statistic covariances with normalizing flows

We explore the ability of normalizing flow (NF) generative models to reproduce weak-lensing summary statistics when trained on a set of cosmological simulations. Our analysis focuses on how accurately NF models recover the mean, standard deviation, and covariance of key statistics derived from convergence ($κ$) maps: The angular power spectrum $C_{\ell}$, probability density function, and Minkowski functionals of weak lensing convergence $κ$-maps. We test two scenarios for training: (1) on the data vectors and (2) on the full $κ$-maps. In both cases, the NF models reproduce the mean and variance of the target statistics within percent-level accuracy. However, the accuracy of the off-diagonal elements of the covariance matrix is underestimated by up to $\sim25\%$. We study several mitigation strategies and find that data augmentation and training with noisy fields help improve covariance recovery to $\mathcal{O}(5\%)$ on power spectrum statistics. Our study demonstrates that while the means and variances of weak lensing statistics can be well modeled by NF, covariances can be significantly underestimated if mitigation strategies are not applied. We present this test as a rigorous diagnostic of generative-model fidelity.

astro-ph.CO↗
arXiv2026

Collider Spin Tomography with Missing Neutrinos

Missing neutrinos need not destroy collider spin tomography. We formulate the visible measurement under kinematic ambiguities arising from invisible particles as a coarse-grained positive-operator-valued measure on the production spin density matrix. We show that information loss is governed by the null space of the resulting visible-data map, not by the number of kinematic solutions. In $e^+e^-\toτ^+τ^-\toπ^+π^-+ν\barν$, the twofold ambiguity leaves only the antisymmetric spin-correlation combination $C_{nr}-C_{rn}$ unidentifiable, while the differential production rate and the remaining fourteen spin coefficients are identifiable. For practical reconstruction under kinematic ambiguities, we develop a self-consistent fixed-point unfolding method using only visible data, without assuming a theoretical production template. Closure tests in Standard Model and anomalous tau-dipole benchmarks show that the method reproduces the truth-level differential production rate and all identifiable spin coefficients, whereas the usual flat average over kinematic folds gives significantly biased reconstructions. When a nontrivial null space is present, the reconstructed identifiable subspace together with positivity yields controlled ranges for concurrence and the CHSH parameter.

hep-ph↗