arXiv subjects

Hui Wu

Publications and source records attributed to Hui Wu.

At least 19 recordsLinked to original sources

arXiv2026

Optical Detection of Coefficient of Thermal Expansion via Frequency Tracking at Cryogenic Temperatures

Accurate characterization of the coefficient of thermal expansion (CTE) in nanoscale thin films is critical for predicting residual stress and ensuring the reliability of modern micro-electromechanical systems and advanced semiconductor architectures. The primary limitation is that traditional methods depend on precise thickness measurements or are restricted to crystalline materials. In this work, we present an in-situ, non-destructive optical methodology to determine the CTE of suspended nanofilms by tracking the temperature-dependence of their vibrational frequency. This approach eliminates thickness measurements entirely and works for amorphous films where X-ray diffraction (XRD) fails. We experimentally validate this technique across a broad temperature range, evaluating membranes in their native microstructure state without requiring additional destructive patterning or complex sample preparation. This non-contact optical method achieves uncertainty below 5.3 * 10-8 K-1, substantially improving upon conventional thin-film CTE metrology.

physics.app-ph↗

arXiv2026

Tunable turning point of the frequency--temperature response in suspended bilayer SiN_x/Al nanofilms

Non-invasive thermometry is highly desirable in cryogenic experiments to avoid the self-heating inherent in conventional sensors. In previous work, temperature was probed via the vibrational eigenfrequency of a suspended single-layer SiNx nanofilm. However, the turning point of the frequency-temperature curve limited the usable measurement range. Moreover, the slow thermal response confined measurements to a stepwise protocol with only discrete temperature points. Here, we show that these limitations can be overcome with a suspended bilayer SiNx/Al nanofilm, whose temperature-dependent vibrational eigenfrequency provides a monotonic, continuous, and repeatable temperature readout. The frequency-temperature dependence is governed by the mismatch between the effective coefficients of thermal expansion (CTE) of the bilayer and the substrate, which makes the turning point tunable through the thickness ratio of the two layers. Over the 90 K to 300 K range, we characterize three bilayer nanofilms with different Al layer thicknesses and show how the turning point shifts with the thickness of the Al layer. For the sample with 72 nm Al, the turning point is pushed beyond 300 K, yielding a monotonic negative temperature coefficient over the entire 90 K to 300 K range. Combining a thin Al layer on the SiNx film with improved thermal anchoring of the chip shortens the thermal response to below 1 s, enabling continuous frequency tracking at a ramp rate of 3 K/min. We also demonstrate that the nanofilm with a 72 nm Al layer achieves a repeatability better than 150 mK above 175 K, and the frequency resolution of 1 Hz corresponds to a temperature resolution of ~5 mK.

cond-mat.mes-hall↗

arXiv2026

Quantum Imaging via Kurtosis-Difference Weighted Covariance on 2D Camera

Camera-based quantum imaging detects spatially correlated photon pairs from spontaneous parametric down-conversion (SPDC). Conventional covariance methods typically require tens of thousands of frames to extract weak correlations from noise. While thick crystals can increase photon flux, they generate photon pairs from multiple emission positions within the crystal, producing multiple correlation centers with complex pairing geometries. In addition, conventional covariance methods assume a single pre-selected correlation center and cannot fully exploit these distributed correlations. We demonstrate that kurtosis difference, a fourth-order statistic measuring tail similarity, effectively discriminates correlated pixel pairs even when correlation coefficients remain low. Weighting covariance by an exponential function of absolute kurtosis difference can select symmetric pixels while preserving true coincidences. This kurtosis weighting automatically identifies correlated pairs within a broad search region and accommodates multiple pairing geometries without requiring precise correlation center calibration. At 5000 frames, our method yields a contrast-to-noise ratio (CNR) exceeding 7, whereas standard covariance remains below 2. Compared with standard covariance, the method reduces the acquisition time by 40-fold and could enable practical quantum imaging in sparse correlated-photon regimes.

quant-ph↗

arXiv2026

Knowledge boundary probing and demand-guided intervention for LLM-based power system code generation

Large language models (LLMs) can turn grid-analysis requests into executable programs for power-system simulation, but utilities and research laboratories often require on-premise deployment. In this setting, first-pass failures frequently arise at an API-knowledge boundary, through hallucinated functions, misused parameters, and mishandled result tables. We present PowerCodeBench, a parameterised benchmark generator released as a frozen 2,000-task suite for pandapower, and a deployment-time workflow that requires no weight updates. Documentation-driven L0-L3 probes produce per-model API profiles for diagnosis, model comparison, documentation allocation, and backend calibration. A query-side demand estimator selects layered API evidence before generation, while execution feedback routes targeted repair. Across ten open-weight LLMs (1.5B-480B) and four mid-tier APIs, the validation-enabled workflow raises scalar-match accuracy by 32-56 percentage points after up to three repair rounds relative to an unassisted first pass, for every model of at least 7B and every API. Open-weight models in the 70B-120B range reach the four-vendor mid-tier accuracy range under matched no-tool conditions. Selective injection approaches the full-layer reference using 41% of its prompt tokens. Among model-item pairs passing numerical checks under both workflows, engineering review confirms the requested analysis in 88% of full-workflow outputs versus 66% under plain repair. Round-0 pilots on OpenDSS and PyPSA motivate staged onboarding from broad retrieval at cold start to calibrated selective injection. Measured throughput, latency, energy, and allocated GPU memory establish a practical on-premise serving envelope.

cs.SE↗

arXiv2026

Conditional Dynamical Systems for Image Generation

Image generation is dominated by deep generative models on GPUs, whose computational and energy costs raise sustainability concerns. Emerging non-von Neumann substrates, including quantum, compute-in-memory, photonic, and thermodynamic platforms, promise greater efficiency, yet most work ports conventional neural network architectures onto them and mainly accelerates operations such as matrix multiplication. This does not fully exploit a native capability of many hardware substrates: relaxation toward low-energy states can itself compute at little cost. Building on this primitive, we develop dynamical systems for image generation based on Ising spins and Kuramoto oscillators. The internal states of both models evolve under dynamics with an explicit Lyapunov energy, and a compact decoder renders the final states as images. For conditional generation, we introduce local conditioning: programmed interactions remain fixed, and one class-dependent bias per node reshapes the landscape, avoiding additional reprogramming and calibration costs. On CIFAR-10, the Ising and Kuramoto models reach clean-FID scores of $6.40$ and $5.37$, respectively, with $4096$ nodes. With a CMOS-compatible hardware realization, the dynamical-system core is projected to cost ${\sim}110.6$~nJ per image, and the full computation, even with the decoder on a GPU, outperforms all GPU baselines in latency and energy. These results suggest that energy-descending dynamics on non-von Neumann substrates can serve as generative computation, offering an efficient path beyond GPUs.

cs.ET↗

arXiv2026

QAM: Quadratic-Accurate Checkpoint Merging via Sequential Consistency

Saved checkpoints record states along a training trajectory, but generally do not determine the updates at states that would be visited under a different schedule. We study how accurately these checkpoints can reconstruct the endpoint of a sequential reference with prescribed update strengths. Under a common local transition model, two checkpoint-index moment conditions characterize all convex merges that agree with this reference through second order. We then prove an information limit that for nondegenerate profiles, no algorithm using only a fixed-length gradient-descent (GD) history with step size $h$ can achieve $o(h^3)$ endpoint error uniformly over a fixed class of smooth, strongly convex losses. The lower bound follows from two losses with identical GD checkpoint histories but sequential reference endpoints separated by $Ω(h^3)$. \textbf{Quadratic-Accurate Merging} (QAM) achieves a matching uniform $O(h^3)$ endpoint error bound. Its explicit coefficients also define the unique profile-dependent merge that exactly matches the sequential GD reference across all fixed quadratic objectives. Across two public Adam checkpoint trajectories (SmolLM3-3B and OpenEuroLLM-Prelude-9B), three windows and three profiles per model, and 15 tasks, QAM shows mixed results for short windows and broader advantages over \textbf{Warmup-Stable and Merge} (WSM) for longer windows. Matched-moment GSM8K diagnostics further show that local consistency alone does not fully determine downstream scores. These results characterize the reconstruction limits of saved histories, provide a coefficient rule that attains the optimal rate, and assess its practical utility.

cs.LG↗

arXiv2026

Do We Really Need KL Divergence for On-Policy Distillation of Large Language Models?

Since the advent of knowledge distillation, KL divergence has been the standard loss in distillation. Recently, on-policy distillation (OPD) has emerged as an efficient post-training paradigm for LLMs. As a distillation method, OPD naturally inherits KL divergence as its standard loss. However, in this work, we find that KL divergence may not be necessary for OPD. We show that simply preserving the update direction is sufficient for effective OPD. As long as the update direction is toward the teacher, OPD works. More precisely, it is not the direction of every token, but the direction of a small subset of tokens where the teacher and student disagree strongly. We first show that simply assigning a reward of (+1) to tokens where the teacher probability is higher than the student probability and (-1) where it is lower, which merely encourages updates toward the teacher, reproduces almost the same training mode as OPD with reverse KL. We further show that only the direction of a small subset of tokens with large teacher-student disagreement is critical, and training works as long as their update direction is toward the teacher, even if other tokens are pulled away from the teacher. And as an application of these findings, we introduce Consensus Multi-Teacher On-Policy Distillation (C-MOPD) to improve Multi-Teacher On-Policy Distillation (MOPD). Unlike MOPD, which routes each sample to a single teacher and may cause capability conflicts across domains, C-MOPD lets every sample be supervised by all teachers. Experiments show that C-MOPD consistently outperforms MOPD on both math and code benchmarks. Our code is available at https://github.com/LeapLabTHU/KL-Free-OPD.

cs.LG↗

arXiv2026

Not All Preferences Deserve Gradients: Understanding Gradient Utility in Offline Reasoning Alignment

Offline preference optimization aligns reasoning models from fixed chosen--rejected pairs, yet standard methods apply gradient updates from every pair regardless of its training value under the current policy. We argue that this uniform treatment is wasteful and potentially harmful. From the perspective of gradient utility, we show that a pair's contribution depends jointly on informativeness and stability. Pair utility drifts as the policy evolves, high-gradient samples can coincide with high-curvature regions, leading to noisy and destabilizing updates, and the most effective supervision comes from stable confident errors where the model is reliably wrong yet curvature remains low. These findings motivate SAGE (Stability-Aware Gradient Efficiency), which maintains difficulty-stratified candidate pools refreshed during training and selects pairs within each pool by a forward-pass signal-to-curvature score. Only pairs with high current utility receive gradient computation; the rest are excluded from backpropagation. On mathematical reasoning benchmarks across multiple model scales, SAGE outperforms full-data and size-matched baselines while producing substantially smoother optimization trajectories.

cs.AI↗

arXiv2026

Integrated Hardware Annealing based on Langevin Dynamics for Ising Machines

Ising machines are non-von Neumann machines designed to solve combinatorial optimization problems (COP) by searching for the ground state, or the lowest energy configuration, within the Ising model. However, Ising machines often face the challenges of getting trapped in local minima due to the complex energy landscapes. Hardware annealing algorithms help mitigate this issue by using a probabilistic approach to steer the system toward the ground state. In this paper, we present a hardware annealing algorithm for Ising machines based on Langevin dynamics, a stochastic perturbation by random noise. Theoretical analysis, system-level design, and detailed circuit design are carried out. We evaluate the performance of the algorithm through chip-level simulation using a standard 65-nm CMOS technology to demonstrate the algorithm's efficacy. The results show that the proposed hardware annealing algorithm effectively guides the system to reach the ground state with a probability of 86.5%, significantly improving the solution quality by 97.5%. Further, we compare the algorithm with state-of-the-art hardware annealing methods through behavioral-level simulations, highlighting its improved solution quality alongside a 50% reduction in time-to-solution.

cs.AR↗

arXiv2026

Delay-induced chimera transitions via mode selection in a multiplex FitzHugh Nagumo network

We investigate delay-induced collective dynamics in a two-layer multiplex FitzHugh Nagumo network with nonlocal intra layer coupling and delayed inter layer interactions. While delay effects are often treated as secondary, we show that deterministic inter-layer delay alone can act as a control mechanism for spatial coherence. Through systematic numerical simulations, we observe a clear transition as the delay parameter increases: fragmented incoherence evolves into chimera-like partial coherence, and eventually into a coherent traveling-wave state. This transition is consistently captured by spatial snapshots, space-time plots, and mean phase velocity profiles. To explain this behavior, we analyze the stability of spatial Fourier modes and show that the delay term introduces a mode-dependent exponential factor in the characteristic equation. This term induces non-monotonic changes in modal stability, effectively acting as a mode-selection mechanism: intermediate delays selectively destabilize a subset of modes, producing chimera-like coexistence, while larger delays suppress incoherent modes and restore global coherence. Our results demonstrate that inter-layer delay provides a simple and robust mechanism for controlling pattern formation in multiplex excitable networks, offering new insight into delay driven synchronization phenomena.

physics.bio-ph↗

arXiv2026

From Intent to Evidence: A Categorical Approach for Structural Evaluation of Deep Research Agents

Deep Research Agents (DRAs) aim to answer complex questions by searching the web, checking evidence, and synthesizing conclusions across heterogeneous sources. We introduce a category-theoretic framework for evaluating and improving such agents. The framework treats deep research as a structured mapping from user intent to evidence-grounded conclusions, making retrieval traces, cross-source alignment, and final synthesis explicit. Guided by this view, we derive a mechanism-aware benchmark of 296 bilingual questions. The benchmark targets four structural skills central to real research: following multi-hop evidence chains, verifying claims across sources, re-ordering fragmented information, and rejecting unsupported assumptions. We evaluate 16 frontier systems with human verification and find that these structural tasks remain highly challenging: the best system reaches only 19.9% average accuracy. The results show that strong agents can sometimes reorganize evidence and detect false premises, but still struggle with long-horizon synthesis and intersection-heavy verification. Beyond evaluation, the same theory also leads to practical system improvements. We instantiate theory-guided interventions such as tracked search, which preserves retrieval traces, and category tools, which add explicit verification and synthesis steps. These interventions yield measurable gains in API-based deep research systems. Our work therefore provides both a challenging benchmark and concrete design guidance for building more reliable research agents.

cs.LG↗

arXiv2026

Delay-Induced Stability Transitions in Directed Signed Consensus Networks

We study delay-induced transitions in consensus dynamics on signed networks with a ring topology. The proposed model is formulated as a system of delay differential equations incorporating both cooperative and antagonistic interactions, as well as heterogeneous time delays. We perform a stability analysis by deriving the associated characteristic equation and examining the real parts of its eigenvalues. It is shown that the stability of the consensus state depends critically on the magnitude of the delays. In particular, increasing time delays may destabilize the system and induce transitions from consensus to bounded non-convergent behavior or instability. A phase diagram in the parameter space is constructed to identify different dynamical regimes. Numerical simulations validate the theoretical results and illustrate the delay-induced transitions. Such delay-induced transitions have also been reported in various biological and engineered systems, highlighting the universal role of time delays in shaping collective dynamics.

math.DS↗

arXiv2026

Housing Decisions under Mobility Risk: A Stochastic Threshold Approach

We develop a stochastic free-boundary model of housing tenure decisions in markets with high mobility risk, such as areas near military installations. Housing prices and rents follow correlated diffusion processes, and households face an uncertain relocation horizon. We derive a closed-form characterization of the optimal buy-versus-rent boundary in terms of the price-to-rent ratio. The model highlights how mobility risk reduces the value of ownership by shortening the effective holding period and increasing uncertainty. As a result, identical price-to-rent ratios can imply different optimal decisions across locations. Numerical illustrations show how variations in volatility and relocation intensity shift the threshold, providing a structural interpretation of observed cross-market heterogeneity. The framework offers a tractable tool for analyzing housing decisions under uncertainty and for interpreting price-to-rent ratios in high-mobility environments.

math.OC↗

arXiv2026

GRAFITE: Generative Regression Analysis Framework for Issue Tracking and Evaluation

Large language models (LLMs) are largely motivated by their performance on popular topics and benchmarks at the time of their release. However, over time, contamination occurs due to significant exposure of benchmark data during training. This poses a risk of model performance inflation if testing is not carefully executed. To address this challenge, we present GRAFITE, a continuous LLM evaluation platform through a comprehensive system for maintaining and evaluating model issues. Our approach enables building a repository of model problems based on user feedback over time and offers a pipeline for assessing LLMs against these issues through quality assurance (QA) tests using LLM-as-a-judge. The platform enables side-by-side comparison of multiple models, facilitating regression detection across different releases. The platform is available at https://github.com/IBM/grafite. The demo video is available at www.youtube.com/watch?v=XFZyoleN56k.

cs.CL↗

arXiv2026

Representations of the modular shifted super Yangian $Y_{1|1}(σ)$

Let $Y_{1|1}$ be the Yangian associated to the general linear Lie superalgebra $\mathfrak{gl}_{1|1}$, defined over an algebraically closed field $\mathbbm{k}$ of characteristic $p>2$. In this paper, we classify the finite dimensional irreducible representations of the restricted super Yangian $Y_{1|1}^{[p]}$ and the restricted truncated shifted super Yangian $Y_{1|1,\ell}^{[p]}(σ)$.

math.RT↗

arXiv2026

BioChemInsight: An Online Platform for Automated Extraction of Chemical Structures and Activity Data from Patents

The automated extraction of chemical structures and their corresponding bioactivity data is essential for accelerating drug discovery and enabling data-driven research. Current optical chemical structure recognition tools lack the capability to autonomously link molecular structures with their bioactivity profiles, posing a significant bottleneck in structure-activity relationship analysis. To address this, we present BioChemInsight, an open-source pipeline that integrates DECIMER Segmentation with MolNexTR for chemical structure recognition, GLM-4.5V for compound identifier association, and PaddleOCR combined with GLM-4.6 for bioactivity extraction and unit normalization. We evaluated BioChemInsight on 181 patents covering 15 therapeutic targets. The system achieved an average extraction accuracy of above 90% across three key tasks: chemical structure recognition, bioactivity data extraction, and compound identifier association. Our analysis indicates that the chemical space covered by patents is largely complementary to that contained in established public database ChEMBL. Consequently, by enabling systematic patent mining, BioChemInsight provides access to chemical information underrepresented in ChEMBL. This capability expands the landscape of explorable compound-target interactions, enriches the data foundation for quantitative structure-activity relationship modeling and targeted screening, and reduces data preprocessing time from weeks to hours. BioChemInsight is available at https://github.com/dahuilangda/BioChemInsight.

q-bio.QM↗

arXiv2026

MARA: A Multimodal Adaptive Retrieval-Augmented Framework for Document Question Answering

Retrieval-based multimodal document QA aims to identify and integrate relevant information from visually rich documents with complex multimodal structures. While retrieval-augmented generation (RAG) has shown strong performance in text-based QA, its extensions to multimodal documents remain underexplored and face significant limitations. Specifically, current approaches rely on query-agnostic document representations that overlook salient content and use static top-k evidence selection, which fails to adapt to the uncertain distribution of relevant information. To address these limitations, we propose the Multimodal Adaptive Retrieval-Augmented (MARA) framework, which introduces query-adaptive mechanisms to both retrieval and generation. MARA consists of two components: a Query-Aligned Region Encoder that builds multi-level document representations and reweights them based on query relevance to improve retrieval precision; and a Self-Reflective Evidence Controller that monitors evidence sufficiency during generation and adaptively incorporates content from lower-ranked sources using a sliding-window strategy. Experiments on six multimodal QA benchmarks demonstrate that MARA consistently improves retrieval relevance and answer quality over existing SOTA method.

cs.IR↗

arXiv2025

InfinityStar: Unified Spacetime AutoRegressive Modeling for Visual Generation

We introduce InfinityStar, a unified spacetime autoregressive framework for high-resolution image and dynamic video synthesis. Building on the recent success of autoregressive modeling in both vision and language, our purely discrete approach jointly captures spatial and temporal dependencies within a single architecture. This unified design naturally supports a variety of generation tasks such as text-to-image, text-to-video, image-to-video, and long interactive video synthesis via straightforward temporal autoregression. Extensive experiments demonstrate that InfinityStar scores 83.74 on VBench, outperforming all autoregressive models by large margins, even surpassing some diffusion competitors like HunyuanVideo. Without extra optimizations, our model generates a 5s, 720p video approximately 10x faster than leading diffusion-based methods. To our knowledge, InfinityStar is the first discrete autoregressive video generator capable of producing industrial level 720p videos. We release all code and models to foster further research in efficient, high-quality video generation.

cs.CV↗