arXiv subjects

Qiang Fu

Publications and source records attributed to Qiang Fu.

At least 19 recordsLinked to original sources

arXiv2026

A regularized lattice Boltzmann model for the solid-liquid phase change

In this study, a regularized lattice Boltzmann method is proposed for solid-liquid phase-change simulations. The governing total-enthalpy equation is solved within a regularized lattice Boltzmann (RLB) framework, which enables accurate tracking of the solid-liquid interface while effectively suppressing numerical diffusion. In addition, the multi-relaxation-time RLB (MRT-RLB) model with standard lattices is developed to solve the flow field. To enhance computational efficiency, an efficient multi-GPU (graphics processing unit) implementation based on CUDA C (Compute Unified Device Architecture) is developed, which is particularly suited for computationally intensive three-dimensional phase-change simulations. The proposed model is validated against a series of one-, two-, and three-dimensional benchmark problems. The results demonstrate that the present solver provides a robust, accurate, and efficient framework for simulating solid-liquid phase-change heat transfer.

physics.flu-dyn↗

arXiv2026

Ancestor Lifting of Quasi-Cyclic Additive Codes and Certified Quantum Codes

Additive codes over $\Ffour$ that are self-orthogonal under the trace inner product correspond to binary quantum stabilizer codes. We study 1-generator $(1,2)$-quasi-cyclic (QC) additive codes and their exact quantum distances. For composite lengths $n=\ell m$, an ancestor-lifting theorem transports dual words of the folded ancestor to the length-$n$ code. A two-layer decomposition for $\ell=2$ and a CRT splitting for odd $\ell$ yield dimension laws and explain equality, escape, and mixed-branch behavior in the certified examples. A palindromic criterion ensures symplectic self-orthogonality, while reversal pairing and a symplectic Gram rank formula describe broader construction families. Exact distances are established by information-set decoding screening, Brouwer--Zimmermann computation on the additive dual, and independent witness verification. We certify twenty-one record-breaking binary quantum codes across twenty entries of Grassl's code table, with twenty-eight certified members in total. Propagation improves thirty-four further entries, and explicit constructions are supplied for three table-tying GV entries marked Code Missing Construction. Six additional candidates are reported only with rigorous upper bounds and heuristic screening evidence. The results connect divisor-lattice structure in additive quasi-cyclic codes with reproducible certification of exact quantum distances.

cs.IT↗

arXiv2026

Learning to Be Monitored: Causal Bounds and Intervention-Aware Post-Training Recipes for Chain-of-Thought Monitorability

Chain-of-thought (CoT) monitoring aims to detect potentially harmful behaviors in large language models (LLMs) from their CoTs. Intervention-based evaluations measure this property by asking whether a monitor can detect the role of an intervention from the CoT when the targeted behavior occurs, typically using minimal-criterion sensitivity ($\mathsf{TPR_{mc}}$) proposed by~\citep{guan2025monitoring}. However, a target behavior observed under intervention is not necessarily caused by it, and we show that this metric can overestimate how often a monitor detects target behaviors actually caused by the intervention. Specifically, we establish $\mathsf{TPR_{mc}}$ as an upper bound on this causal detection rate and derive a complementary lower bound, which provides a tighter evaluation framework for CoT monitorability. With this framework, we study whether LLMs can be trained to make their CoTs easier to monitor. We first construct a training dataset of paired control--intervention tasks spanning multiple intervention types. Using the dataset, we design and apply three intervention-specific training methods based on off-policy self-distillation, reinforcement learning, and on-policy self-distillation to train separate variants of each base model. Across various model-monitor configurations, all three methods generally improve CoT monitorability over the corresponding base models under our extended evaluation framework. Our code is available at https://github.com/xinyue-lxy/learning-to-be-monitored.

cs.LG↗

arXiv2026

Disentangling forced and internal components in observed climate evolution using spatiotemporal deep learning

Attribution of observed historical climate evolution to externally forced change and internal variability is fundamental to understanding climate change and evaluating climate models. Yet such attribution remains challenging because the forced response is intertwined with internal variability in observations. Here, we develop a convolutional long short-term memory network (ConvLSTM) to disentangle these two components. Trained and evaluated using large-ensemble simulations from multiple climate models, ConvLSTM consistently outperforms existing model-based, statistical, and machine-learning approaches. Applying ConvLSTM to observations reveals an evolving spatial pattern of the forced response and substantial contributions from internal variability to recent multidecadal climate trends. For several key climate features whose attribution remains debated, ConvLSTM indicates that internal variability amplified the forced reversal of Atlantic interhemispheric temperature asymmetry, dominated the observed La Niña-like warming pattern over the equatorial Pacific and the associated precipitation decline over southwestern North America since 1980, and modulated Arctic amplification around its nearly constant threefold forced value. Comparison between the ConvLSTM-estimated and model-simulated forced responses reveals broadly consistent patterns but also important discrepancies, with climate models simulating stronger Northern Hemisphere warming, weaker subtropical high intensification, and smaller land precipitation declines. Overall, ConvLSTM provides a robust framework for attributing historical climate evolution and evaluating the fidelity of climate models

physics.ao-ph↗

arXiv2026

Hermitian Hulls of Cyclic Codes and Exact Dual Distances of Extended Linear Codes

Let $\mathcal{C}$ be an $[n,k,d]$ linear code over $\F_{q^2}$ with parity-check matrix $H$. The rank $r=\Rank(HH^{\dagger})$ measures how far the Hermitian dual $\mathcal{C}^{\perp_h}$ is from being Hermitian self-orthogonal, and it is the least number of coordinates that must be adjoined to $\mathcal{C}^{\perp_h}$ to make it Hermitian self-orthogonal; equivalently, the Hermitian hull $\mathcal{C}\cap\mathcal{C}^{\perp_h}$ has dimension $(n-k)-r$. In this paper, we develop a structure theory of the Hermitian hull of a cyclic code and apply it to the construction of quantum stabilizer codes. For a cyclic code with defining set $T\subseteq\mathbb{Z}_n$, we show that $r$ is additive over the orbits of the involution $τ(ν)=-qν$ on the $q^2$-cyclotomic cosets and satisfies a parity constraint. A parameter dictionary exhibits $r$ as a gauge trading one-for-one against the code length, and a complementation duality shows that the complement of $T$ realises the logical-qubit count of the quantum code. The extreme values of $r$ are settled by closed subset-sum criteria, and the $τ$-invariant defining sets are shown to be the unique locus on which the Hermitian dual is again cyclic. On the distance side, we determine the Hermitian dual distance of an $r$-fold extension of an arbitrary linear code exactly, and we prove that the admissible extension vectors lie in the image of $HH^{\dagger}$, which reduces the search space from $q^{2(n-k)}$ to $q^{2r}$. Applied to cyclic, quasi-cyclic and BCH codes, the construction improves $23$ best-known parameters in Grassl's code tables, one of them provably optimal.

cs.IT↗

arXiv2026

Multi-pulse accumulation of gas molecular coherence enables gigahertz ultrafast frequency conversion

Frequency conversion of ultrafast lasers is fundamentally constrained by the trade-off between pulse energy and repetition rate. Here, we demonstrate a frequency conversion mechanism in which molecular coherence accumulates across ultrafast pulse trains in gas-filled hollow-core fibers. Unlike conventional nonlinear interactions initiated by individual high-energy pulses, this mechanism relies on the collective buildup of coherent molecular oscillations driven by successive pulses. Using this approach, we achieve Raman frequency conversion at repetition rates up to 3 GHz with nanojoule pulse energies. The results demonstrate that accumulated molecular coherence can govern nonlinear frequency conversion, revealing a domain of light-matter interaction whose physical limits remain largely unexplored.

physics.optics↗

arXiv2026

Affine Schur--Weyl theory for loop Semigroups and polynomial representations

Classical Schur--Weyl duality plays a fundamental role in the representation theory of general linear and symmetric groups. In this paper, we develop an affine Schur--Weyl theory for the Laurent polynomial loop semigroup. Let $\hat G_K(n)$ denote the Laurent polynomial loop semigroup inside the loop group $GL_n(K((t)))$ over a field $K$, and let $\hat S_K(n,r)$ be the affine Schur algebra. The natural action of $\hat G_K(n)$ on the affine tensor space $Ω_K^{\otimes r}$ induces an algebra homomorphism $K\hat G_K(n)\rightarrow \hat S_K(n,r).$ We prove that this homomorphism is surjective for every field $K$ satisfying $|K|>r$. Furthermore, we establish the double centralizer property for affine Schur algebras over unique factorization domains for all $n\geq2$. For $n\geq2$ and any field $K$ with $|K|>r$, this yields the corresponding centralizer property for $\hat G_K(n)$. In contrast, for $n=1$ and $r\geq2$, the corresponding natural homomorphism is not surjective over any field. For $n\geq 2$, we use the double centralizer theorem to determine the center of the affine Schur algebra and, when $K$ is a field with $|K|>r$, show that $\hatζ_{r,K}$ maps the center of $K\hat G_K(n)$ onto it. Over an algebraically closed field $K$ of arbitrary characteristic, we prove that every finite-dimensional irreducible $\hat S_K(n,r)$-module is a tensor product of evaluation modules. Moreover, we obtain explicit character formulas expressing these irreducible characters as products of characters of irreducible classical Schur algebra modules. Finally, we study polynomial representations of $\hat G_K(n)$ over algebraically closed fields $K$, and classify its finite-dimensional irreducible polynomial representations. For finite fields $\mathbb F_q$ with $q>r$, this yields a classification in the defining-characteristic setting.

math.RA↗

arXiv2026

ReTeach: Building a Self-Teacher through Multi-Round Reflection and Retry

Self-distillation can improve reasoning without a separately trained, more capable teacher, but its effectiveness depends on how the self-teacher gains an advantage over the student. Conditioning the teacher on reference answers or solutions can provide such an advantage, but this information may be unavailable. Reflection offers a way to derive explicit error diagnoses and revision guidance from self-generated attempts, yet existing reflection-based methods often combine it with reference information, rich task feedback, or persistent memory. We introduce ReTeach, a Reflective self-distillation framework that constructs its self-Teacher through multi-round reflection and retry using only self-generated attempts and outcome-level verification. Starting from an unsuccessful student rollout, the teacher alternates explicit reflection with renewed attempts until success or the retry budget is exhausted, without reference answers or solutions, external diagnostic feedback, or cross-example memory. Each failed retry informs subsequent reflection, while successful correction provides outcome-level evidence for the potential utility of the resulting teacher context. An outcome-aware selection and weighting strategy distinguishes initially correct, reflection-corrected, and unresolved examples, assigning separate weights to their category-normalized distillation losses. Through on-policy distillation, the student matches the teacher's context-conditioned token-level predictive distributions at prefixes of its own rollouts, transferring the benefits of iterative correction while retaining single-pass inference. Across six benchmarks spanning mathematical reasoning, science question answering, and tool use, ReTeach improves average accuracy over GRPO by 1.39 percentage points.

cs.AI↗

arXiv2026

The climb problem for $4$-general sets in PG(n,4)

A $4$-general set of PG(n,q) is a point set with no four coplanar,and Mb{n}{q} is the largest such size. The climb problem asks, at each $n$, whether the constructions of Pavese (2025) can be improved by one point. We show that the answer is governed by the code tables: $PG(M-1,q)$ contains an $N$-point $4$-general set exactly when a projective $[N,N-M,\ge 5]_q$ code exists. Over $F_4$ this settles the first two rungs: Mb{3}{4}=5 and Mb{4}{4}=11, each with two independent proofs, so Pavese's constructions are optimal at both. Over $F_5$ the same reduction gives Mb{4}{5}=12, one more than the near-MDS lower bound. At $n=5$ the tables fall silent; we state the verified position $21\le Mb{5}{4}\le 30$ rather than the tabulated $\le 29$, which rests on a private communication. Any $22$-point $4$-general set of $PG(5,4)$, if one exists, shares at most $12$ points with Pavese's extremal $21$-point set; the proof of this rigidity statement is a finite computation, with certificates in Appendix~A.

math.CO↗

arXiv2026

Low Latency Gaze Tracking via Latent Optical Sensing

We present a real-time gaze tracking system that directly acquires task-relevant latent features using a fully passive optical encoder. Instead of forming and processing full-resolution images, our approach leverages a microlens array with a co-designed binary chromium mask to perform spatially multiplexed optical encoding, producing a compact set of measurements sufficient for gaze estimation. By integrating sensing and feature extraction in the optical domain, the proposed system eliminates the need for high-bandwidth image readout and substantially reduces computational overhead. The encoded measurements are captured by a 4 x 4 phototransistor array and mapped to gaze direction using a lightweight neural network. Our proof-of-concept prototype enables an end-to-end sensing-to-inference latency of 3.4 ms, outperforming published research systems. We demonstrate the effectiveness of our approach on both simulated and real-world data, achieving competitive gaze estimation accuracy while significantly reducing end-to-end latency compared to conventional camera-based pipelines. The proposed system establishes a new operating point in the accuracy-latency-compute trade-off for latency- and resource-constrained gaze tracking. This work highlights the potential of task-driven optical sensing for ultra-low-latency, computationally efficient human-computer interaction systems.

cs.CV↗

arXiv2026

Quantum current algebra ${\bf U}(\frak{gl}_n[t])$: canonical bases and polynomial representations

We introduce a quantum deformation $\mathbf{U}(\mathfrak{gl}_n[t])$ of the universal enveloping algebra of the current algebra $\mathfrak{gl}_n[t]$, realized as a Hopf parabolic subalgebra of quantum affine $\mathfrak{gl}_n$. Unlike the Yangian---the standard quantization of the current algebra---our algebra admits a canonical basis. We give a BLM-type realization of $\mathbf{U}(\mathfrak{gl}_n[t])$ via certain subalgebras of affine quantum Schur algebras, and then construct canonical bases for the modified quantum current algebra $\dot{\mathbf{U}}(\mathfrak{gl}_n[t])$ and for its finite dimensional irreducible graded modules. Moreover, we prove a rigidity theorem: every finite dimensional polynomial irreducible module for quantum affine $\mathfrak{gl}_n$ remains irreducible when restricted to ${\mathbf U}_{\mathbf v}(\frak{gl}_n[t])$ (the specialization of ${\mathbf U}(\frak{gl}_n[t])$ at a non-root-of-unity complex number $\mathbf{v}$); conversely, every finite dimensional affine polynomial irreducible ${\mathbf U}_{\mathbf v}(\frak{gl}_n[t])$-module extends uniquely to a polynomial irreducible module for quantum affine $\mathfrak{gl}_n$. More generally, we classify all finite dimensional polynomial irreducible ${\mathbf U}_{\mathbf v}(\frak{gl}_n[t])$-modules by $\mathcal Q(n)\timesΛ^+(n)$. As an application, we establish a bijection between the isomorphism classes of finite dimensional polynomial irreducible modules of ${\bf U}_{\mathbf v}(\frak{gl}_n[t])$ and those of the Yangian $Y(\mathfrak{gl}_n)$. Thus ${\bf U}(\mathfrak{gl}_n[t])$ admits canonical bases for its modified form and irreducible graded modules, and its finite dimensional polynomial irreducible modules are in correspondence with those of the Yangian.

math.QA↗

arXiv2026

Affine quantum Schur--Weyl duality

Let $\mathpzc K$ be an arbitrary commutative ring containing an invertible element $\varepsilon$. Let ${\mathcal H}_{\!\vartriangle\!}(r)_{\mathpzc K}$ be the extended affine Hecke algebra of type $A$ with Hecke parameter $\varepsilon$, let $Ω_{\mathpzc K}^{\otimes r}$ be the affine tensor space, and let ${\mathcal S}_{\!\vartriangle\!}(n,r)_{\mathpzc K}$ be the corresponding affine quantum Schur algebra. We first prove that the natural right action of ${\mathcal H}_{\!\vartriangle\!}(r)_{\mathpzc K}$ on $Ω_{\mathpzc K}^{\otimes r}$ is always faithful. Assume further that $\mathpzc K$ is a field of characteristic $0$ and that $\varepsilon$ is not a root of unity. We prove that, for any $n\geq 2$, the natural algebra homomorphism $ξ_r:{\mathcal H}_{\!\vartriangle\!}(r)_{\mathpzc K}\rightarrow\operatorname{End}_{{\mathcal S}_{\!\vartriangle\!}(n,r)_{\mathpzc K}}(Ω_{\mathpzc K}^{\otimes r})^{\mathrm{op}}$ is an isomorphism. This proves Conjecture~3.8.8 of \cite{DDF}. As an application, we prove the conjecture formulated in \cite[5.2.4]{DDF} concerning the center of the affine quantum Schur algebra. We also prove that ${\mathcal S}_{\!\vartriangle\!}(n,r)_{\mathpzc K}$ is left and right Noetherian whenever $\mathpzc K$ is a Noetherian commutative ring, which verify a conjecture in \cite[Rem. 1.7]{DY}.

math.QA↗

arXiv2026

High-capacity computing with self-rectification nonlinear optical neural processor

Artificial intelligence (AI) and neural networks have driven groundbreaking innovations across numerous disciplines. Optical computing offers the promise of unprecedented speed and energy efficiency in the post-Moore era; however, achieving efficient, practical nonlinear activation using all-optical approaches remains a challenge. Here, we present an optical nonlinear neural processing unit (ONNPU) that implements all-optical nonlinear activation through a self-rectification mechanism. The ONNPU architecture perfectly imitates the structure of digital neural networks, enabling seamless integration with the established deep learning ecosystem. We benchmark ONNPU across nine diverse tasks spanning decision, regression and generation, including accuracies of 98.07% on MNIST and 93.54% on Fashion-MNIST. When integrated into a 201-million-parameter Vision Transformer, ONNPU achieves 82.4% top-1 accuracy on full ImageNet classification (1,000 categories); when integrated into a 117-million-parameter decoder-only Transformer, ONNPU enables short-form story generation that outperforms GPT-2. By addressing more complex and diverse deep learning tasks, ONNPU paves the way toward practical optical machine intelligence, unleashing significant potential for high-performance optical computing.

physics.optics↗

arXiv2026

Centers of quantum Schur superalgebras from Hecke algebras

We study the center of the quantum Schur superalgebra $\mathcal{S}_v(m|n,r)$ associated with the general linear Lie superalgebra $\mathfrak{gl}_{m|n}$. Using the super Schur--Weyl duality due to Mitsuhashi between the quantum supergroup $U_v(\mathfrak{gl}_{m|n})$ and the Hecke algebra $\mathcal{H}_v(\mathfrak{S}_r)$, we transfer two known bases of the center of $\mathcal{H}_v(\mathfrak{S}_r)$, namely the Geck--Rouquier basis and the Jones basis, to the center of $\mathcal{S}_v(m|n,r)$. This yields two distinct bases for $\mathscr{Z}(\mathcal{S}_v(m|n,r))$, indexed respectively by the symmetrized hook set $H^{\vee}(m|n,r):=H(\min(m,n)\mid \max(m,n),r)$ and by the full $(m|n)$-hook set $H(m|n,r)$. Our approach relies on a detailed analysis of the ring $Λ_{m|n}$ of doubly symmetric polynomials satisfying $f|_{x_m=t=-y_n}$ independent of $t$, and of its power-sum bases.

math.QA↗

arXiv2026

Quantum Locally Repairable Codes from Negacyclic and Repeated-Root Cyclic Codes over Small Fields

Quantum locally recoverable codes (qLRCs), introduced recently by Golowich and Guruswami, allow any single-qudit erasure to be recovered from a small set of other qudits. Most known constructions require a large alphabet. We systematically investigate qLRCs obtained, via the CSS construction, from classical constacyclic codes over small fields $\Ff_q$ with $q\in\{2,3,4,5,7\}$. First, we prove that a nonzero dual-containing $λ$-constacyclic code exists only when $λ^2=1$, so that negacyclic and (repeated-root) cyclic codes exhaust the constacyclic route to qLRCs. Second, we show that the locality of a constacyclic code equals the minimum distance of its dual minus one, and we give a simple purity criterion for the resulting quantum codes. Third, we show that odd-like duadic codes whose splitting is given by $μ_{-1}$ yield pure qLRCs; specializing to $q$-ary quadratic residue codes of prime length $p\equiv 3 \pmod 4$ gives an infinite family of pure qLRCs with unbounded minimum distance and certified locality. Finally, by means of concrete computations, we obtain a classification of qLRCs from cyclic, negacyclic, and repeated-root cyclic codes of moderate lengths, which contains the first binary qLRCs from repeated-root cyclic codes and many parameter sets that cyclic codes cannot attain.

cs.IT↗

arXiv2026

Kimi K3: Open Frontier Intelligence

We introduce Kimi K3, a 2.8T parameter Mixture-of-Experts model with 104 billion activated parameters, native vision capabilities, and a 1-million-token context window. Kimi K3 is built on Kimi Delta Attention and Attention Residuals, which improve information flow across sequence length and model depth. Together with Stable LatentMoE, which effectively activates 16 of 896 routed experts per token, and refined training and data recipes, these advances yield an approximately 2.5x improvement in overall scaling efficiency over Kimi K2. Post-training highlights reinforcement learning across general, agentic, and coding domains and multiple reasoning-effort levels, enabling compositional generalization and robust long-horizon execution. At 2.8T scale, Kimi K3 is supported by infrastructure advances in multiple areas: algorithm-system co-design for KDA, perfectly balanced expert-parallel training with efficient memory management, million-token agentic RL with persistent rollout and sandbox states, and deployment innovations. Extensive evaluations show that Kimi K3 achieves frontier-level performance across long-horizon coding, agentic, knowledge, reasoning, and vision tasks. While its overall performance still trails the most powerful proprietary models, namely Claude Fable 5 and GPT-5.6 Sol, Kimi K3 consistently outperforms other open and proprietary models evaluated in our suite. We release the full Kimi K3 model weights to facilitate future research and accelerate the broader deployment and adoption of frontier intelligence.

cs.CL↗

arXiv2026

Scalable LLM Agent Tool Access in the Cloud

LLM agents increasingly rely on tool calling to act on external systems, and the Model Context Protocol (MCP) has quickly become its de facto interface. Operating MCP at cloud scale, however, becomes difficult. On the tool provider side, legacy services are not directly callable through MCP; the rapid protocol development also creates ongoing compatibility cost. On the agent side, the number of accessible tool is limited by the LLM context window and inference overhead; mounting a large tool set increases token usage and inference latency and can reduce task success rate. Moreover, for stateful MCP backends with multiple replicas, preserving session affinity increases client-side complexity. We present a cloud-scale gateway system for MCP service. It breaks the direct-connect model on the data plane and offloads legacy service integration, consolidating incompatible MCP variants, access control, tool recommendation, and session-aware routing to the gateway. Hybrid retrieval sustains 98% Top-15 recall; it scales agent tool access to 3,000+ with high tool selection accuracy, and reduces tool selection time by $8.9\times$ and token usage by $23.8\times$, with low per-call overhead, stable under scale-out. Finally, we share the lessons learned from deploying the gateway system in production.

cs.DC↗

arXiv2026

Integrating Power Electronics-based Energy Storages to Power Systems: A Review on Dynamic Modeling, Analysis, and Future Challenges

The integration of power electronics-based energy storage systems (PEESs) into power systems introduces potential instabilities. This study reviews efforts in dynamic analysis of both AC and DC power systems integrated with PEESs, covering dynamic modeling, analysis methods, and potential instability risks. Major conclusions are drawn as: 1) Simplified models of PEESs have been widely used for dynamic analysis of power systems. However, it may cause "error aggregation" as the scale of PEESs increases, leading to mistakes in results, which induces significant concerns. 2) Traditional stability mechanism analysis methods remain effective for single grid-connected PEES and large-scale PEESs with parallel and series connections. However, they are inadequate for PEESs with distributed connections. To fill in this gap, an idea of mechanism analysis based on "dynamic reconstruction" is proposed. 3) Potential instability risks caused by PEESs integration may differ from those caused by renewable energy integration due to differences in functional controls and bidirectional power flow. However, comprehensive investigations in this regard are lacking and require significant attention. To ensure the stable operation of power systems with increasing integration of PEESs, significant challenges are summarized in the end, providing inspirations for future studies.

eess.SY↗