arXiv subjects

Yang He

Publications and source records attributed to Yang He.

At least 19 recordsLinked to original sources

arXiv2026

From Generation to Simulation: How Far Are World Models from Being True Simulators?

With the rapid progress of diffusion models and large-scale video generation, generative world models are increasingly expected to replace traditional simulators, including physics engines, game engines, and reinforcement-learning environments. Yet the remaining distance from generation to simulation lacks a systematic assessment. We present a capability-based study using an external yardstick: eight capabilities of a traditional simulator, namely asset construction, physics engine, interaction, controllability, stability, state feedback, diversity, and evaluation metrics. We trace three main technical routes--latent dynamics, video generation, and joint-embedding prediction--and map exactly 200 representative works published from 2018 to June 2026 onto these capabilities. Our analysis shows that world models have achieved functional substitution in interaction and controllability for specific scenarios, but remain short of traditional simulators in formal guarantees of physical laws, structured state feedback, and reproducible long-horizon evolution. State feedback is the most neglected cross-route shortcoming: only 6 of 163 implementation papers expose a runtime interface for querying entity states or physical parameters. We identify six research directions: formalized physics, a unified action interface, first-class state feedback, long-horizon stability, downstream-utility evaluation, and cross-route hybridization. Project page: https://github.com/AtongWang/world-model-simulators

cs.AI↗

arXiv2026

Analytic Boundary Terms for Arbitrary Crystal Geometries and Direct-Sum Evaluation of Madelung Constants in Triclinic Lattices

The direct-sum evaluation of Madelung constants is complicated by the conditional convergence of lattice sums, which gives rise to a shape-dependent boundary term. In this work, we present, for the first time, a closed-form analytic expression for this boundary term that is valid for arbitrary crystal geometries. For general triclinic lattices, this boundary term maps exactly onto the electrostatic potential generated by a set of uniformly charged parallelograms. In addition, we demonstrate that the residual finite-size correction for a crystal of characteristic size $p$ decays as $(2p+1)^{-2}$. Building on these results, we develop a robust direct-sum method for the accurate computation of Madelung constants in arbitrary triclinic lattices and validate its effectiveness through explicit calculations on representative Bravais lattices.

cond-mat.other↗

arXiv2026

Active Passivation Tunes Hotspot Locations in GaN Transistors with In Situ Thermal Mechanical Visualization

Efficient thermal dissipation has become critical in emerging electronic devices. However, most existing studies have primarily focused on engineering heat dissipation pathways, largely overlooking the intrinsic behavior of the heat source itself. We demonstrate an active passivation technology that proactively tunes hotspot locations in GaN transistors. By adjusting the active passivation layer length, the hotspot is shifted from the gate edge to the drain-side AP edge, establishing a clear one-to-one spatial correlation. In-situ thermal-mechanical visualization via micro-Raman thermography, combined with multi-physics electro-thermal-mechanical simulations, directly captures the spatial redistribution of both temperature and thermal stress profiles. Electrical analysis confirms that this hotspot migration is driven by the spatial shift of the peak electric field and localized Joule heating. This proactive heat-source tuning strategy provides critical design guidelines for power electronics.

physics.app-ph↗

arXiv2026

SR-Agent: An Experience-Driven Agentic Framework for Post-Ranking Strategy Refinement in E-Commerce Recommendation

User experience is a first-class objective in industrial e-commerce recommender systems (RS). Post-ranking strategies, which govern diversity, similarity, and exposure over a ranked list, are widely deployed in industrial RS for their simplicity and low serving cost. However, as the online recommendation environment evolves continuously, these statically configured strategies gradually become stale, thereby degrading the user experience. Refining them typically relies on manual inspection, diagnosis, and updates, making it slow, costly, and difficult to scale or reuse. Although recent LLM-based agents (e.g., RecUserSim, SimUSER, and Self-EvolveRec) offer promising directions, none of them close the full loop of automated, self-evolving strategy refinement. To bridge this gap, we introduce SR-Agent, which, to the best of our knowledge, is the first agentic framework deployed to refine post-ranking strategies in industrial RS. SR-Agent unifies three components: (i) a UserSim agent that applies inspection skills to surface user-perceived bad cases; (ii) an Analysis agent that consolidates recurring bad cases into structured, reusable diagnoses; and (iii) a constrained Strategy Refinement Harness that maps diagnoses to typed and bounded actions, gated by a four-stage reward pipeline with reversible rollback. Deployed on the Kuaishou e-commerce platform, SR-Agent continuously runs this refinement loop and, in a one-month online A/B test, increases order volume by 0.71%, browsing depth by 0.34%, and clicked-category diversity by 0.48%, while markedly shortening the refinement cycle and lowering operational cost.

cs.AI↗

arXiv2026

Neutron-source fidelity for laser-driven D--D lithium-blanket tritium-breeding tests

Compact deuterium--deuterium (D--D) neutron sources can provide controllable irradiation fields for lithium-blanket studies, although their broad joint energy and angle distributions differ from the conventional $2.45$~MeV isotropic representation. We couple particle-in-cell (PIC) simulations of target-normal-sheath-accelerated deuterons with a thick-target $D(d,n)^{3}$He source model and Monte Carlo neutron transport. For natural lithium, the seven two-dimensional sources change tritium production per source neutron by $-2.5\%$ to $+54.1\%$ relative to the ideal source. The matched three-dimensional calculation gives an increase of $43.5\%$ and lowers the corresponding ratio from $1.5406$ to $1.4350$. Source substitutions show that the difference is predominantly spectral, since the real spectrum alone gives a factor of $1.4199$, while using the real neutron emission directions in place of isotropic emission adds only a further factor of $1.0106$ in the three-dimensional case. The real spectrum lowers the $^{6}$Li contribution by $6.9\%$, but the accessible $^{7}$Li$(n,Xt)$ response exceeds this loss. Enrichment to $90\%$ $^{6}$Li keeps the total change within $\pm1.5\%$. In the matched three-dimensional converter and blanket calculation, direct $D(d,p)$T production is $0.8458$ tritons per source neutron and accounts for $98.1\%$ and $86.9\%$ of the combined production for natural and enriched lithium, respectively. High-density polyethylene moderation raises tritium production by about one order of magnitude but first weakens and then reverses the increase in blanket tritium production. The analysis quantifies source-model effects in compact breeding tests.

physics.plasm-ph↗

arXiv2026

ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog

Despite growing automation, turning a paper into a coherent poster, talk video, and blog piece often remains a labor-intensive last mile. Recent systems increasingly generate multiple dissemination formats, but a practical workflow must also keep the outputs editable in native tools and bound into one navigable deliverable for revision and reuse. We present ResearchStudio-Reel, a native-editable dissemination workspace that binds its three artifacts into one interactive deliverable at the experience level, implemented as five skills executable in Claude Code and Codex: one shared extractor, three editable artifact generators, and one interactive convergence layer. A shared asset bundle feeds a PowerPoint poster and video deck, plus a bilingual Word blog; rather than re-rendering the paper into a fourth format, Paper2Reel converges these already-produced artifacts at the experience level, binding poster regions, video segments, and blog passages into one interactive viewer. Artifact-specific release checks make this delivery contract testable, and Paper2Poster additionally uses a measured-fill loop. On the Paper2Poster benchmark, our Claude Code configuration achieves the best scores among automated systems on all three aesthetic sub-criteria and the best or tied-best scores on two of three information sub-criteria. Under two VLMjudges, it exceeds the authors' posters in average aesthetics (3.56 vs. 3.03) and wins on overall quality on 74 and 95 of the 100 papers under the two judges. The full pipeline additionally packages the native-editable source artifacts and their aligned viewer. Project is available at https://aka.ms/ResearchStudio

cs.CV↗

arXiv2026

ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes

Large language models have made research ideation increasingly accessible, yet effective idea development requires more than generating candidate directions. Researchers must ground a problem in current literature, identify meaningful bottlenecks, differentiate from existing solutions, and evaluate risks before committing to implementation. We present ResearchStudio-Idea as a reusable skill suite for this first mile of research ideation. The suite includes Paper-Search, a standalone multi-source literature search skill; Scoop-Check, a standalone prior-art collision checker for novelty claims; and IdeaSpark, the end-to-end skill that composes evidence grounding, pattern-guided generation, collision retrieval, audit, and idea-card rendering into one workflow. IdeaSpark is constructed from a corpus of 1,947 machine learning conference papers collected from ICLR, ICML, and NeurIPS between 2021 and 2025, including Oral papers, a separately tracked high-citation subset, and rejected submissions. Analysis of these outcomes reveals 31 recurring ideation sub-patterns, consolidated into 15 reusable ideation patterns. Each pattern is operationalized as a structured card containing research contexts, bottleneck types, differentiation strategies, supporting precedents, and common failure modes. Given a research problem and an evidence bundle, IdeaSpark evaluates evidence readiness, reconstructs the surrounding research context, identifies unresolved bottlenecks, selects relevant patterns, instantiates one candidate direction, retrieves potentially conflicting prior work, and performs outcome-informed auditing. This workflow transforms reusable ideation patterns into traceable research proposals. Blind automated-judge evaluations show that IdeaSpark consistently produces stronger research proposals than no-skill and generic-skill baselines while maintaining competitive novelty.

cs.AI↗

arXiv2026

A superconducting qutrit link beyond the qubit limit

Superconducting microwave links have enabled deterministic state transfer and remote entanglement between qubits, but deterministic links have so far operated with an effectively two-dimensional transmitted Hilbert space. Here we demonstrate a superconducting qutrit link between two independently packaged nodes connected by a microwave channel. Each node combines a transmon qutrit, a transmission resonator, and a tunable Purcell-filter interface, allowing the two remote microwave-photon interfaces to be matched in both frequency and bandwidth. We implement two transition-selective photon-mediated operations that transfer the $|e\rangle$ and $|f\rangle$ qutrit components in distinct temporal modes of the same channel. We tomographically characterize arbitrary qutrit-state transfer, obtaining a mean transferred-state fidelity of 83.68% and a qutrit process fidelity of 77.12%, exceeding both the classical qutrit-transfer benchmark and the best possible average fidelity of an effective qubit channel used to transmit an arbitrary qutrit. Using partial-transfer operations, we reconstruct a remote two-qutrit state with negativity 0.730, a tomography-inferred dense-coding capacity of 2.273 bits, and a tomography-inferred Collins-Gisin-Linden-Massar-Popescu (CGLMP) parameter $I_3=2.332$, all beyond the corresponding qubit or local bounds. These results demonstrate a superconducting microwave link that uses the native three-level structure of transmons as a genuine high-dimensional communication resource.

quant-ph↗

arXiv2026

The Finite Coulomb Lattice Sum: A Resolution of Conditional Convergence through Exact Shape and Size

This work examines conditionally convergent Coulomb lattice sums under periodic boundary conditions. The recently developed finite lattice sum cleanly decomposes the series into three distinct components: a periodic bulk term $\nu_{\rm pbc}$, a shape-dependent non-periodic boundary term $\nu_{\rm b}$, and a finite-size correction term $\nu_{\rm corr}$. This rigorous formulation explicitly parameterizes the geometry of a finite lattice by its exact shape and size and takes an effective pairwise form. We analyze it in detail and compare it with various derivations of lattice sums in the literature. Perspectives on future applications are discussed, including analytical developments for arbitrarily shaped crystals and numerical mesh-type algorithms for condensed matter simulations.

physics.chem-ph↗

arXiv2026

Multi-objective Bayesian optimisation of a double-layer target for quasi-monoenergetic TNSA protons

We carry out a six-parameter multi-objective Bayesian optimisation of a carbon--hydrogen double-layer target for target-normal-sheath proton acceleration. The campaign consists of 80 two-dimensional EPOCH simulations with the laser amplitude $a_0$, pulse duration $\tau$, carbon-layer thickness $L_1$, hydrogen-layer density $N_2$, hydrogen-layer thickness $L_2$ and hydrogen-layer radius $r_p$ as input variables. Each final proton spectrum is scored by the peak energy, the charge fraction inside a $\pm10\%$ peak-energy window and the charge in that window. Among the Pareto-set evaluations, the cases with peak energies between 64 and 71 MeV occur near $a_0=30$, $\tau=45$ fs, $L_1=0.3\,\mu{\rm m}$, $L_2=30$ nm and $r_p=0.15\,\mu{\rm m}$. Along this branch, increasing $N_2$ raises the in-window charge and increases the bandwidth. The small rear-layer radius keeps the proton source within the flat central region of the transverse sheath field, where the accelerating field is nearly uniform. A 3D calculation is performed for the intermediate-density case $N_2=11.85\,n_c$, which balances bandwidth and in-window charge along this branch. The corresponding 2D spectrum has $E_{\rm peak}=67.4$ MeV and $\Delta E/E=18.8\%$, whereas the 3D spectrum has $E_{\rm peak}=34.1$ MeV and $\Delta E/E=7.0\%$. The lower 3D peak energy and narrower bandwidth are associated with an earlier decay of the rear-sheath field and an earlier saturation of the proton peak energy, and the quasi-monoenergetic peak is retained in 3D.

physics.plasm-ph↗

arXiv2026

Bayesian optimization of double-pulse temporal shaping for enhanced target-normal-sheath proton acceleration under fixed laser energy

Splitting an ultrashort drive pulse into a weak leading pulse and a strong main pulse is known to raise the energy of protons accelerated by the target-normal-sheath-acceleration (TNSA) mechanism, because the leading pulseforms a preplasma that increases the absorption of the main pulse. The allocation of energy between the two pulses and their temporal separation are coupled control parameters, and under a fixed total energy they have not been optimized jointly in a systematic way. We address this problem with two-dimensional particle-in-cell simulations driven by Bayesian optimization. Treating the prepulse energy fraction $r$ and the interpulse delay $\Delta t$ as free parameters under a fixed total energy, a campaign of 32 simulations, of which 16 are Sobol-initialized and 16 adaptively selected, locates an optimum at $r\approx0.07$ and $\Delta t\approx234$~fs. The proton cutoff energy increases from 7.7~MeV for the single pulse to 17.7~MeV at the optimum, a gain of about 130\%. The optimum is asymmetric with only about 7\% of the energy in the leading pulse. At the optimum the laser absorption rises from 4.84\% to 20.09\%, the bulk hot-electron temperature from 1.20 to 1.94~MeV, and the time-integrated rear sheath field by a factor of about 1.7. The optimum lies on a broad plateau in $\Delta t$, which relaxes the timing tolerance required in an experiment.

physics.plasm-ph↗

arXiv2026

VIA-SD: Verification via Intra-Model Routing for Speculative Decoding

Speculative decoding (SD) addresses the high inference costs of LLMs by having lightweight drafters generate candidates for large verifiers to validate in parallel. Existing draft-verify methods use binary decisions: accept or fully recompute. Yet we find that many rejected tokens can be verified correctly by a slim submodel derived from the full verifier via intra-model routing, instead of the full verifier. This motivates our slim-verifier to handle tokens requiring moderate verification resources, reducing expensive large-model calls. We propose Verification via Intra-Model Routing for Speculative Decoding (VIA-SD), a multi-tier framework using a routed slim-verifier. Draft tokens are processed hierarchically: direct acceptance for high-confidence cases, slim-verifier regeneration for medium-confidence cases, and full-model verification for uncertain cases. Across four representative tasks and multiple model families, VIA-SD reduces rejection rates by 0.10-0.22 and delivers 10-20% speedups over strong SD baselines, while achieving 2.5-3x acceleration over non-drafting decoding. Moreover, VIA-SD is compatible with existing SD frameworks without modifying their training procedures. Our results suggest multi-tier SD as a general paradigm for scalable and efficient LLM inference. Project page: https://zju-xyc.github.io/VIA-SD-Project-Page/

cs.CL↗

arXiv2026

From 2D Grids to 1D Tokens: Reforming Shared Representations for Multimodal Image Fusion

Multimodal image fusion aims to integrate complementary information from different modalities into a fused image that preserves rich local details while maintaining globally consistent appearance. Existing approaches build shared representations on 2D feature grids, which excel at modeling local structures but offer limited leverage over image-level global appearance factors. To balance these objectives, we introduce a compact 1D token interface based on a frozen pretrained image tokenizer for modeling non-local appearance/base factors. Rather than using the tokenizer as a reconstruction backbone, our design uses the 1D token space as a global carrier while retaining the 2D spatial pathway for local structure restoration. Specifically, we introduce Selective Token Editing (STE), which sparsely updates/replaces a small set of critical tokens, providing a lightweight mechanism to steer global appearance coherence while keeping the fusion backbone unchanged and avoiding extra losses. Experiments on four commonly used benchmarks show that our method achieves the best overall performance, with consistent, multi-metric improvements in both global coherence and local fidelity. Project page: https://zju-xyc.github.io/1D-Fusion-Project-Page/

cs.CV↗

arXiv2026

Accuracy-Configurable Floating-Point Multiplier Design for SRAM-Based Compute-in-Memory

Digital Compute-in-Memory (DCiM) reduces data movement and has become a promising solution for energy-efficient edge AI. However, most existing DCiM frameworks still primarily target integer or fixed-point arithmetic, and provide limited support for compiler-integrated and accuracy-configurable floating-point computation. Directly integrating conventional IEEE 754 floating-point units into dense SRAM-based DCiM arrays, however, incurs high area and power overhead. To address this challenge, this work presents an accuracy-configurable floating-point multiplier integrated into the OpenACM framework for SRAM-based DCiM. An exact IEEE~754-compliant multiplier is first implemented as a baseline, and a mantissa-segmentation-based approximate multiplier is then proposed to reduce hardware cost while preserving numerical fidelity. Post-layout results show up to 69% logic area reduction and 72% power savings over exact floating-point designs without delay overhead. Evaluations on image processing tasks and ResNet-18 inference further demonstrate negligible accuracy degradation. These results indicate that compiler-integrated approximate floating-point multiplication is a practical approach for enabling efficient and configurable floating-point support in SRAM-based DCiM systems. The Floating-Point Multiplier is available on https://github.com/ShenShan123/OpenACM

cs.AR↗

arXiv2026

A Surrogate Model for Proton Spectrum Prediction to Map Transitions in Laser-Ion Acceleration

We present a physics-guided, decoupled dual-branch surrogate model to predict continuous proton energy spectra from laser-driven ion acceleration. Integrating a $\beta$-VAE for spectral feature extraction with a parallel multi-layer perceptron for scalar boundary enforcement, the framework achieves a predictive accuracy of $R^2 = 0.94$ for the maximum cutoff energy and $R^2 = 0.94$ for the total particle flux, with a median per-sample spectral $R^2 = 0.985$ (in $\log_{10}$ space) across the full 2000-bin energy distribution. The model incorporates uncertainty quantification via deep ensembles, serving as a quantitative probabilistic diagnostic tool with calibration errors below 6.2\%. Within the 1D longitudinal framework, the surrogate reproduces spectral signatures consistent with the transition from Target Normal Sheath Acceleration (TNSA) to the volumetric heating dynamics of Relativistically Induced Transparency (RIT) and Breakout Afterburner (BOA) regimes, as validated against kinetic diagnostics from 1D particle-in-cell simulations. This approach establishes a computationally efficient baseline for future multi-fidelity optimization and provides an engine for closed-loop parameter control in high-repetition-rate laser facilities.

physics.plasm-ph↗

arXiv2026

Programmable spectral symmetries in an anisotropic quantum Rabi simulator

The quantum Rabi model captures fundamental aspects of light--matter interaction, where symmetry dictates both spectra and dynamics. Over the past years, experiments have explored many of its nonperturbative properties, but have mostly focused on the isotropic limit, where rotating and counterrotating processes are locked together, leaving the broader symmetry landscape largely unexplored. Here we realize a programmable anisotropic quantum Rabi model in a superconducting processor, with independent control of the rotating and counterrotating couplings $(g_1,g_2)$ and of a transverse bias $\varepsilon$. Continuous anisotropy tuning, combined with a duality mapping, gives access to the full parameter space from the Jaynes-Cummings to the anti-Jaynes-Cummings limits. In the deep-strong-coupling regime, we show that anisotropy reconstructs the spectrum and turns complete collapse-revival dynamics into incomplete revivals even near degeneracy. With adiabatic state preparation and joint tomography, we resolve an anisotropy-induced ground-state parity switch, a crossing that has no analogue in the isotropic model. We further observe selective tunnelling associated with hidden symmetry in biased Rabi models and track its anisotropic displacement within the same device. These results establish a controllable route to engineering nonperturbative light--matter Hamiltonians, where symmetry, spectrum, and dynamics can be programmed independently.

quant-ph↗

arXiv2026

DeepTool: Scaling Interleaved Deliberation in Tool-Integrated Reasoning via Process-Supervised Reinforcement Learning

Tool-Integrated Reasoning (TIR) extends LLM capabilities by leveraging external environments. However, existing methods lack the deliberation during sequential tool invocation required for strategic planning and self-correction. While RL mitigates this, conventional approaches for Tool-Integrated Reasoning are hindered by sparse outcome-based rewards, failing to supervise intermediate reasoning steps and tool invocations. To address this, we propose DeepTool, a novel framework that scales deliberate thinking within the interleaved process of thinking, action, and observation at each turn. In DeepTool, we first introduce a synthesis pipeline that evolves extended thinking into interleaved trajectories, integrating adversarial perturbations to ensure robustness and self-correction. Secondly, we devise Process-Supervised Reinforcement Learning based on GRPO, which utilizes an Action-Centric Process Reward to reinforce intermediate interleaved thinking and enforce precise tool invocation at every turn. Extensive experiments demonstrate that DeepTool achieves superior performance, boosting Qwen2.5-7B significantly across six benchmarks (e.g., AIME24: 3.2% -> 40.4% and HMMT25: 0.0% -> 28.6%). Furthermore, the token cost-effectiveness analysis confirms the utility of interleaved thinking, demonstrating DeepTool's optimal balance between performance and token efficiency.

cs.AI↗

arXiv2026

Fusion of Integrable Defects and the Defect $g$-Function

We study exact defect $g$-functions for integrable line defects in two-dimensional integrable quantum field theory and use them to probe defect fusion. We consider three settings: fusion of purely transmitting topological defects, fusion of non-topological defects with reflection and transmission, and fusion of a defect with an integrable boundary. For topological defects, the separated logarithmic $g$-function is additive, and the fusion limit is controlled by the multiplicative composition of transmission factors. For non-topological defects, separation-dependent phases in the Bethe-Yang equations produce oscillatory finite-size effects, while the fused defect is described by effective reflection and transmission amplitudes. In the Ising examples studied here, fusion involving non-topological defects lowers the finite localized contribution to the entropy, whereas topological defect-boundary fusion leaves it unchanged.

hep-th↗