arXiv subjects

Yan Wang

Publications and source records attributed to Yan Wang.

At least 73 records · Page 4Linked to original sources

arXiv2026

DreamQAS: Learning a Decision-Useful World Model for VQE-Efficient Quantum Architecture Search

Reinforcement-learning-based quantum architecture search (RL-QAS) repeatedly invokes a variational quantum eigensolver (VQE) after each gate addition even though circuit transitions and action legality are known. DreamQAS preserves these exact dynamics and learns only expensive post-VQE feedback through a recurrent ensemble that predicts a frontier-relative feedback score without requiring the exact ground-state energy, enabling uncertainty-controlled multi-step imagination. Under a common 15,000-episode budget and frozen evaluation, DreamQAS has the lowest reported mean error among RL methods on all five main molecular tasks. At fine-error targets reached by all seeds of DreamQAS and a matched non-imaginative control, it uses 1.6-2.0 times fewer real VQE calls on four tasks. Holding LiH-4q feedback-model weights fixed, its imagined-policy actor attains 0.073 mHa, versus 4.280 mHa and 4.434 mHa for greedy and beam deployment. Learned-transition and end-to-end predictor controls further show that preserving exact circuit structure and using feedback through policy learning are both important. Counterfactual action-ranking improves throughout training on all five probed tasks, while ensemble disagreement improves risk-coverage over random rejection on three tasks. DreamQAS therefore learns decision-useful feedback for QAS without modeling already-known circuit dynamics or requiring the exact ground-state energy.

cs.LG↗

arXiv2026

MomentBA: Second-order Spatial Moments for Anisotropic Correspondence Uncertainty in Differentiable Bundle Adjustment

Most existing visual odometry (VO) systems treat feature correspondences as deterministic measurements or assign uniform uncertainty, ignoring the inherent localization ambiguity of different observations. However, correspondence uncertainty is often anisotropic due to image structures such as edges, repetitive patterns, and motion blur, which can significantly affect geometric optimization. In this work, we propose MomentBA, a geometry-aware bundle adjustment framework that derives anisotropic correspondence uncertainty from second-order spatial moments of local similarity responses. Instead of introducing additional covariance prediction networks, the proposed method directly converts matching response distributions into interpretable covariance estimates and incorporates them into bundle adjustment as correspondence-specific information matrices for uncertainty-aware residual weighting. Furthermore, the proposed formulation is integrated into a differentiable optimization framework, establishing a direct connection between correspondence uncertainty and geometric estimation. Experiments on the EuRoC MAV and TartanAir v1 Hard datasets demonstrate that MomentBA improves monocular visual odometry accuracy compared with existing feature-based and learning-based approaches. The proposed anisotropic covariance model achieves lower rotational errors and more robust trajectory estimation than fixed and isotropic uncertainty models, validating the effectiveness of geometry-induced uncertainty modeling for challenging visual environments.

cs.CV↗

arXiv2026

Prediction-Market Seed Capital Recovery from Noise-Dominant Flow

Automated prediction markets require sponsors to prefund liquidity before observing order flow, creating a financing challenge at launch. We study whether nonnegative charges conditioned on observable payoff direction can improve recovery of this prefunded capital while limiting their effect on informed participation. We develop Seed Capital Flow (SCF), a direction-conditioned levy, in a stylized binary cost-function market with informed and liquidity-motivated traders. When order composition differs across directions, SCF concentrates the permitted fee burden on the direction with relatively more liquidity-motivated flow, whereas a uniform fee spreads it across both directions. Under a sufficiently tight common retention constraint, this allocation yields higher expected recovery capacity and can make additional liquidity choices financially viable. Synthetic numerical audits examine robustness to alternative flow patterns, stochastic arrivals, and label misspecification. The results characterize a mechanism-design tradeoff rather than an empirical prediction: the market remains prefunded, recovery is expected rather than guaranteed, and the analysis is limited to an opening-cohort setting.

cs.CE↗

arXiv2026

Energy Accuracy Is Not Enough: A Structure-Aware Benchmark and Evaluation Protocol for Quantum Architecture Search

Quantum architecture search for molecular ground-state estimation is commonly evaluated through energy accuracy, which does not describe circuit cost or the physical properties of the prepared state. We introduce HamQASBench, a structure-aware benchmark comprising eleven molecular Hamiltonians of up to fourteen qubits, selected using Hamiltonian and target-state properties and supplied with exact references. Its evaluation protocol combines energy accuracy and success rates with reference-relative circuit cost, local entropy profiles for non-degenerate targets, and state identification within degenerate ground subspaces. Experiments with five methods spanning four search paradigms reveal differences hidden by energy-only comparisons. On a near-product instance under the shallow search budget, the best outputs of all five methods reach chemical accuracy while using between two and sixty-two gates. Equal-energy outputs on a degenerate instance occupy distinct spin components. Local entropy profiles distinguish inaccurate outputs and show that entangling-gate counts need not reflect realized entanglement. Across the molecular instance ladder, product-state outputs can meet or miss chemical accuracy, motivating interpretation of success alongside target-state properties. These results support retaining energy as the task-success criterion while using circuit-cost and state diagnostics for more informative comparisons. The benchmark instances, references, evaluation implementation, and per-run data are released for reuse.

quant-ph↗

arXiv2026

SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking

Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by selecting a small set of context units (tokens or blocks) for each query. Existing trainable methods usually use a lightweight selector to score context units, followed by hard Top-K selection that blocks gradients from the language modeling loss. Consequently, these methods commonly distill layer-wise dense attention distributions. Although this encourages the selector to rank context units by dense attention weights in the original model, the ranking is not directly aligned with their impact on predictions under a fixed attention budget (i.e., the number of attended context units per query), potentially wasting the limited budget on less useful units. To address this misalignment, we propose Simple Attention Sparsification (SAS), a gated sparse attention mechanism that optimizes context ranking end-to-end with the language modeling loss. The key idea is to inject the selector's continuous scores into attention logits during training, allowing the loss to update the selector through standard backpropagation. We identify several choices crucial for this simple design to work well in practice: placing the gate inside the attention softmax in log form, using normalized softmax gates to calibrate historical context against the always-retained current block, and preserving continuous selector scores so the model learns relative priorities rather than only hard selections. To support long-sequence training, we implement a memory-efficient Triton kernel that integrates SAS into FlashAttention-style computation. Across reasoning, long-context understanding, and agentic tasks, SAS consistently outperforms trainable sparse attention baselines across attention budgets, with especially large gains under tight budgets, demonstrating more effective context ranking for downstream tasks.

cs.CL↗

arXiv2026

Moonworks Lunara: Modeling Artistic Intelligence

We formulate \emph{Artistic Intelligence} as exploration driven world realization, leaving space for creative possibility while preserving the semantic, artistic, and compositional structure that must remain true. Moonworks Lunara, a text-to-image model, implements this framework with a novel Diffusion Mixture Transformer architecture. A new training algorithm iteratively evolves the data distribution through informative sample acquisition and targeted injection of human-created art. We benchmark Lunara against seven image-generation models, including FLUX.2-Klein-4B, Qwen-Image (20B), and GPT-Image-1-Mini. With GPT-5.6 Sol as evaluator, Lunara ranks first in \emph{Aesthetic Quality (8.473 vs. 8.457 GPT-Image-1-mini)}, second in \emph{Emotional Resonance}, and remains competitive in \emph{Content Integrity}. A blind human evaluation over the same evaluation set corroborates the automated metrics, ranking Lunara first. It also stays among the strongest models under conventional measures including CLIPScore and LAION Aesthetic Predictor. On GenEval, Lunara achieves competitive performance against a broader set of 16 models, including GPT Image 2 and Seedream 4.0. These results place Lunara at the frontier with Artistic Intelligence while maintaining a sub-10B active-parameter footprint and sub-10-second inference latency. Lunara advances the general visual intelligence frontier by shifting the question from whether models can get images right to how deeply they can interpret meaning and realize it as imaginative, expressive worlds.

cs.CV↗

arXiv2026

Exploring the Enigmatic Chiral Phase Transition of QCD at FiniteTemperature

This paper offers a comprehensive overview of recent technological advancements concerning the critical point of chiral phase transition, with a particular focus on Effective Field. Theories in Quantum Chromodynamics (QCD). It delves into the intricacies of these pivotal points, utilizing theoretical tools to explore associated phenomena.

hep-th↗

arXiv2026

Legendre Transformation under Micro Canonical Ensemble

The Legendre transformation is a crucial tool in theoretical physics, known for its symmetry, especially when applied to multivariate functions. In statistical mechanics, ensembles represent the central focus. Leveraging the dimensionless aspect of Legendre transformation, this paper explores the transformation process from the entropy characteristic function of microcanonical ensembles to the analogous definition of partition function transformation. Additionally, it derives characteristic functions, partition functions, and establishes their interrelations, along with deriving corresponding thermodynamic formulas for various ensembles. This streamlined approach sheds light on the fundamental principles of statistical mechanics and underscores the symmetry inherent in Legendre transformation.

cond-mat.stat-mech↗

arXiv2026

Basic Data Processing of Gravitational Waves

This paper provides a comprehensive guide to gravitational wave data processing, with a particular focus on signal generation, noise modeling, and optimization techniques. Beginning with an introduction to gravitational waves and the detection techniques used by LIGO and Virgo, the manual covers the essentials of signal processing, including Fourier analysis, filtering, and the generation of quadratic chirp signals. The analysis of colored Gaussian noise and its impact on interferometric detectors like LIGO is explored in detail, alongside signal detection methods such as the Generalized Likelihood Ratio Test (GLRT). The paper also delves into optimization techniques like Particle Swarm Optimization (PSO), which can be applied to improve signal estimation accuracy. By providing MATLAB-based implementations, this manual serves as both a theoretical and practical resource for researchers in the field of gravitational wave astronomy.

astro-ph.IM↗

arXiv2026

A Robust Watermark-based Fingerprint Framework for GNNs Ownership Verification

The high training cost of Graph Neural Networks (GNNs) has raised growing concerns regarding model ownership infringement, such as model stealing and unauthorized misuse. To verify model ownership and prevent significant economic losses, two groups of GNN Ownership Verification (OV) methods have been proposed: watermark-based methods and fingerprint-based methods. However, these methods typically face three limitations: (1) the performance degradation of protected models caused by out-of-distribution (OOD) watermark graphs with respect to the training set; (2) the unrealistic assumption that surrogate models have been trained on a watermark-containing training set; and (3) over-reliance on specific output levels for fingerprint extraction. In this paper, we propose a Robust watErMArk-based fingeRprint frameworK for GNNs, named REMARK. REMARK first generates carefully crafted in-distribution watermark graphs that maximize output differences between GNN models, thus mitigating OOD-induced performance degradation. REMARK then extracts robust fingerprints from these output differences to verify GNN ownership, thereby removing the assumptions that surrogate models must be trained on a watermark-containing dataset or expose specific output levels. Extensive experiments across widely used real-world datasets and GNN architectures demonstrate that REMARK achieves state-of-the-art OV accuracy and robustness while preserving the utility of protected models.

cs.LG↗

arXiv2026

T2LSC-Bench: Benchmarking Localized Semantic Control in Text-to-Image Generation

Recent text-to-image models have become increasingly capable of rendering explicit text, but reliable localized text control requires more than generating the correct string. In applications such as product labeling, signage, and interface design, target text should be rendered within a designated text-bearing region without altering the predefined subject identity or surrounding scene semantics. We refer to violations of this requirement as target-text-associated semantic leakage, in which target-text semantics are expressed through non-textual visual content beyond the designated anchor. Existing visual-text benchmarks primarily evaluate readability, spelling accuracy, and layout, leaving this form of semantic leakage largely unexamined. We introduce T2LSC-Bench, a controlled diagnostic benchmark comprising 50 seed subjects and 1,200 prompt cases per model, yielding 7,160 evaluated images across six models. Its factorized design varies semantic relation, scene openness, prompt mode, and language. A dual-branch protocol combines OCR-VLM text verification with structured VLM semantic judgments to measure Text-at-Anchor Accuracy (TAA), Semantic Subject Preservation (SSP), Semantic Leakage Rate (SLR), and Conditional Semantic Leakage Rate (cSLR). Under stress-test conditions, SLR increases from 1.2% to 18.1% and cSLR from 1.3% to 18.2%, whereas TAA decreases only from 91.4% to 90.9%. Anti-leakage prompting reduces SLR from 16.6% to 8.4% without degrading rendering accuracy. Human validation on 420 images shows strong agreement between automatic and adjudicated annotations. These results show that accurate text rendering does not guarantee local containment of target-text semantics.

cs.CV↗

arXiv2026

Real-time heralded non-Gaussian teleportation resource-state generator

Quantum teleportation is a fundamental quantum communications primitive that requires an entangled resource state. In the continuous-variable regime, non-Gaussian entangled resources have been shown theoretically to improve teleportation fidelity compared to Gaussian squeezed vacuum. We experimentally demonstrate a heralded two-mode resource state for non-Gaussian teleportation capable of real-time use. We characterize this state with two-mode homodyne tomography showing it has fidelity $F=0.973\pm 0.005$ with the expected resource state. Real-time use is enabled by a photon-subtraction orchestrator system performing live coincidence detection and outputting low-jitter and low-latency heralding signals. Live collection of real-time quadrature measurements of photon-subtracted states is enabled by the development of a synchronized homodyne detection server, where the orchestrator system queries the server to collect the real-time quadrature samples corresponding to the heralded state. These results demonstrate significant advancement in enabling the use of heralded non-Gaussian states in quantum networking protocols, especially in the context of quantum repeaters, non-Gaussian quantum sensing, and measurement-based quantum computing.

quant-ph↗

arXiv2026

Beyond Dense Adam States: Adaptive Log-Space Quantization for Memory-Efficient Optimizers

Optimizer-state quantization is commonly designed for Adam's dense, parameter-aligned first- and second-moment arrays. This abstraction breaks for memory-efficient optimizers, whose states may be factored, confidence-modulated, or maintained in a projected space, so similar reconstruction error can produce different update error. We formulate optimizer-state quantization as a joint problem over representation, topology, and update semantics. We then introduce Adaptive Log-Space (AL) quantization for non-negative states. AL fits each block's observed nonzero logarithmic interval and reserves a separate code for exact zero, enforcing $q = 0 \Leftrightarrow x = 0$; signed momentum and state precision remain independently selectable. Controlled probes show that adaptive ranges reduce update error and temporal drift, exact-zero reservation preserves dormant states, and state topology constrains useful block granularity. End-to-end language-model training evaluates the resulting policy across dense, factored, confidence, and projected optimizer states. On TinyLlama-1.1B, AL8 with uniform 8-bit momentum reaches 72.90 perplexity versus 73.54 for bitsandbytes 8-bit AdamW, with comparable optimizer-state storage and higher throughput. CAME matches reference-level final perplexity across three seeds when its non-negative states use AL16, while a semantic grouping-and-protection policy closes most of quantized Adafactor's 100K-step late-loss gap. These results make state topology and update semantics first-class design constraints for optimizer quantization.

cs.LG↗

arXiv2026

MIRAGE: Misleading Retrieval-Augmented Generation via Black-box and Query-agnostic Poisoning Attacks

Retrieval-Augmented Generation (RAG) systems enhance LLMs with external knowledge but introduce a critical attack surface: corpus poisoning. While recent studies have demonstrated the potential of such attacks, they typically rely on impractical assumptions, such as white-box access or known user queries, thereby underestimating the difficulty of real-world exploitation. In this paper, we bridge this gap by proposing MIRAGE, a novel multi-stage poisoning pipeline designed for strict black-box and query-agnostic environments. Operating on surrogate model feedback, MIRAGE functions as an automated optimization framework that integrates three key mechanisms: it utilizes persona-driven query synthesis to approximate latent user search distributions, employs semantic anchoring to imperceptibly embed these intents for high retrieval visibility, and leverages an adversarial variant of Test-Time Preference Optimization (TPO) to maximize persuasion. To rigorously evaluate this threat, we construct a new benchmark derived from three long-form, domain-specific datasets. Extensive experiments demonstrate that MIRAGE significantly outperforms existing baselines in both attack efficacy and stealthiness, exhibiting remarkable transferability across diverse retriever-LLM configurations and highlighting the urgent need for robust defense strategies.

cs.CR↗

arXiv2026

Compactness of abundance in asymmetric hypergraph removal lemmas

Fix an integer $r\ge 2$ and a finite simple $r$-uniform hypergraph $F$ with at least one edge and no isolated vertices. An $n$-vertex $r$-graph is $ε$-far from being $F$-free if at least $εn^r$ edges must be deleted to destroy every copy of $F$. A finite $r$-graph $H$ is $F$-abundant if there are constants $c,C>0$ such that every sufficiently large $ε$-far host contains at least $cε^C n^{v(H)}$ labelled copies of $H$. A family is $F$-abundant when one member has this lower bound in each host, although the member may depend on the host and on $ε$, while $c$ and $C$ are common to the family. We prove that every abundant family contains an abundant member. We prove the analogous coloured theorem for $F$-partite hosts containing edge-disjoint part-respecting copies of $F$ such that every vertex lies in at least $εn^{r-1}$ of them. The case $r=2$ yields the coloured and uncoloured graph compactness theorems, answers Question 5.2 of Girão, Hurley, Illingworth and Michel, and proves their Conjecture 5.1 [J. Lond. Math. Soc., 2024]. We also obtain an explicit bound $\lfloor 2r(C+1)\rfloor$ for the order of the non-isolated core of a selected witness. Moreover, we give several applications. For example, we construct translation-invariant linear systems from abundant coloured hypergraphs, obtain a square-root bound for an equation associated with a cycle of bounded length, give a one-sided tester based on one fixed graph when distance from the property gives a polynomial lower bound on distance from being $F$-free, and prove that no algorithm decides whether a family of finite simple graphs enumerated by a Turing machine is $K_3$-abundant.

math.CO↗

arXiv2026

Extremal Families for the Erdős--Kleitman Problem: The Missing Constructions

For integers $n\ge s\ge2$, let $e(n,s)$ be the maximum size of a family $\mathcal F\subseteq2^{[n]}$ with no $s$ pairwise disjoint members. The problem of determining $e(n,s)$, now called the Erdős--Kleitman problem, is closely related to the well-known Erdős matching problem. Frankl and Kupavskii posed a meta-conjecture predicting that the maximum is always attained by a weighted family. Fix $m\ge3$, write $n=ms+c$ with $0\le c 0$, $β=β(m,k)>0$ and an integer $s_0=s_0(m,k)$ such that, for all integers $s\ge s_0$ and all integers $c$ with $0\le c<s$, the only extremal families for $e(n,s)$ are the families $\mathcal H^k(m,s,\ell;A)$ with $A\in\binom{[n]}{a_k}$ whenever $βs^{(k-1)/k}\le c\le αs^{k/(k+1)}$. In particular, this result determines an infinite number of new extremal families for the Erdős--Kleitman problem and verifies the Frankl--Kupavskii meta-conjecture in these ranges. This also provides a quantitative extension of the result of Kupavskii and Sokolov on the extremality of $\mathcal H^1$.

math.CO↗

arXiv2026

An Optimal Bound for Ramsey Goodness of Cycles

For graphs $F$ and $H$, the Ramsey number $R(F,H)$ is the minimum integer $N$ such that every $N$-vertex graph contains $F$ or its complement contains $H$. If $F$ is connected and $|F|\geσ(H)$, a construction of Burr gives $R(F,H)\ge(χ(H)-1)(|F|-1)+σ(H)$, where $σ(H)$ denotes the minimum order of a color class in a proper $χ(H)$-coloring of $H$. Burr proved that this bound is attained for $F=C_n$ when $n$ is sufficiently large. Allen, Brightwell and Skokan conjectured that equality already holds whenever $n\geq |H| χ(H)$, while Haslegrave, Hyde, Kim and Liu subsequently proved it whenever $n\ge C|H|\log^4χ(H)$. Pokrovskiy and Sudakov conjectured that the optimal linear condition $n\geq C|H|$ suffices; this conjecture was also highlighted by Montgomery in his 2026 ICM survey (see Conjecture 9.2). In this paper, we resolve this conjecture by proving that there is an absolute constant $C>0$ such that $R(C_n,H)=(χ(H)-1)(n-1)+σ(H)$ for every nonempty graph $H$ and every $n\ge C|H|$. This gives the first bound linear in $|H|$, and is best possible up to a constant factor. Our proof builds on the framework of Haslegrave, Hyde, Kim and Liu, and combines some new ideas in expansion and switching cycle lengths.

math.CO↗

arXiv2026

Dynamic-Robust Photometric-Semantic Reconstruction for Open-Vocabulary 3D Scene Understanding

The integration of novel view synthesis (NVS) and open-vocabulary segmentation (OVS) has recently yielded powerful feed-forward 3D foundation models. However, their inherent reliance on static-scene assumptions leads to severe misalignment of spatial features in unconstrained dynamic environments. To bridge this critical gap, we propose SPAR, a novel joint semantic-geometric encoding architecture that explicitly isolates transient dynamic noise prior to latent space aggregation. Furthermore, we introduce a dynamic-region-aware end-to-end training paradigm that structurally couples motion estimation with multi-view visual and semantic learning. This unified approach enables the network to inherently resolve motion conflicts and distill multi-view consistent, temporally stable scene representations from dynamic inputs. Extensive experiments on the challenging D-RE10K benchmark demonstrate that SPAR achieves state-of-the-art performance. Our end-to-end approach achieves exceptional novel view synthesis quality, yielding a PSNR of 22.15 dB and 23.33 dB given only 3 and 4 input views respectively. Despite being trained in a self-supervised manner, our model achieves an mIoU of 88.5% for motion mask prediction. Furthermore, our analysis reveals a strong inter-task synergy between photometric scene reconstruction and semantic understanding, where semantic synthesis learning consistently enhances photometric fidelity in novel view rendering. Code will be available at https://github.com/dmucby/SPAR.

cs.CV↗