arXiv subjects

Qinglin Yang

Publications and source records attributed to Qinglin Yang.

At least 19 recordsLinked to original sources

arXiv2026

A compact analytic formula for the one-loop triangle cosmological correlator

We derive a compact analytic formula for the one-loop triangle correlator of conformally coupled scalars in de Sitter space. The result is organised as six leading-singularity prefactors multiplying pure weight-two functions of the six energy variables. It contains forty-two dilogarithms, compared with approximately one hundred and twenty in the previously known closed-form representation, and requires no auxiliary regulator. The dilogarithms occur in Galois-conjugate pairs, making each contribution separately real throughout the physical region. We validate the result numerically. Moreover, we show that its symbol can be derived directly from the dressed integral representation or, independently, bootstrapped from Landau singularities and general consistency conditions. Finally, we show that the correlator (in a suitable normalization) is a Stieltjes function of each squared energy separately and is jointly completely monotone in all six squared energies. These structures suggest a route towards higher-point one-loop cosmological correlators.

hep-th↗

arXiv2026

Federated Prompt Learning: A Unified Framework, Empirical Analysis, and Future Directions

Large language models (LLMs) have become core components of cloud-based intelligent services in academia and industry, yet their training and deployment are hindered by high computational costs, data centralization, and privacy concerns. Federated learning (FL) offers a decentralized training paradigm that enables clients to collaboratively train a learning model without sharing raw data, making it a promising solution for privacy-preserving LLM training and reasoning. This paper presents a comprehensive survey of federated prompt learning (FPL) to review recent advances in integrating the federated learning paradigm and large language models, answering the following research questions: RQ1: The fundamental motivations, characteristics, and enabling technologies of FPL, and how it differs from conventional FL and full-model federated fine-tuning; RQ2: The trade-offs FPL approaches exhibit in performance, communication efficiency, computational overhead, scalability, personalization, and heterogeneity handling; RQ3: The remaining security, privacy, robustness, and system challenges, along with key future research directions. To this end, we systematically examine existing FPL methods across the full model lifecycle: pre-training, fine-tuning, and practical applications, while discussing security, privacy, and robustness issues and summarizing existing defense mechanisms. Finally, we highlight open challenges and future directions, aiming to help readers understand how the insights drive research in FPL.

cs.LG↗

arXiv2026

Embedded Blockchain Infrastructure Management (eBIM): A RISC-V-Empowered Hardware--Software Co-Design Framework Towards Trustworthy Blockchain

Blockchain systems are undergoing a fundamental transition from decentralized ledgers for digital assets to general-purpose trust infrastructures for verifiable computation, decentralized physical resources, and automated infrastructure management. Meanwhile, the limitations of the Blockchain as a Service (BaaS) model stem from a common structural problem: outsourcing control of infrastructure to third-party service providers inevitably involves a systemic surrender of trust, flexibility, and data sovereignty. RISC-V, with its open, modular, and extensible design, provides a general-purpose computing foundation for public blockchains that is open, low-level, compileable, verifiable, and scalable. Inspired by the development and characteristics of eSIM, the embedded Blockchain infrastructure management (eBIM) is defined as a software-hardware collaborative paradigm for blockchain infrastructure management with RISC-V. This study aims to provide a comprehensive survey on eBIM supporting research and technologies, to answer the following research questions (RQs): RQ1 What is eBIM? RQ2 How does eBIM work? RQ3 What can eBIM do? By introducing the concept of eBIM, this paper establishes a foundational reference for researchers, hardware architects, and protocol designers in this rapidly evolving landscape, including cryptographic acceleration, trusted execution environments, zero-knowledge virtual machines, and smart contract execution engines. The prospects of the proposed e-BIM and its future research directions are indicated in this paper.

cs.CR↗

arXiv2026

Bootstrapping two-loop six-gluon amplitudes in QCD

The maximally transcendental, or most complicated, terms of gauge-theory scattering amplitudes have long been singled out, following Lipatov and collaborators, as those parts of a QCD amplitude that most closely mirror maximally supersymmetric Yang--Mills theory. We report on a programme that turns this observation into a practical computational tool. We show that the rational prefactors multiplying the highest-weight special functions of planar QCD amplitudes are governed by four-dimensional leading singularities, which can be classified and evaluated using on-shell diagrams. The resulting prefactors are manifestly conformally invariant and admit compact spinor-helicity expressions that hold for arbitrary multiplicity. Combining this input with the recently established two-loop six-particle function space, we set up a symbol bootstrap and determine, for the first time, the maximal-weight symbol of the planar two-loop six-gluon amplitude in massless QCD, first for the ${-}{-}{+}{+}{+}{+}$ helicity configuration and subsequently for all MHV configurations. The answer is fixed uniquely by physical consistency conditions, requires a reduced alphabet of only $137$ symbol letters, and yields as a byproduct previously unknown two-loop triple-collinear and double-soft splitting functions. We summarise the method and the results, and outline the directions they open up.

hep-ph↗

arXiv2026

Bootstrapping the Four-Point NMHV Stress-Tensor Form Factor

We bootstrap the four-point next-to-maximally helicity-violating (NMHV) form factor of the chiral stress-tensor supermultiplet in planar maximally supersymmetric Yang-Mills theory through three loops at the symbol level. At two loops, an ansatz built from NMHV leading singularities and the relevant five-point one-mass integral function space is fixed uniquely by physical constraints; the resulting ratio function symbol contains 78 letters, all drawn from the 88-letter alphabet previously identified in the four-point MHV sector. At three loops, using this 88-letter alphabet as input and imposing extended Steinmann relations satisfied by the minimally-subtracted hard function, together with other physical constraints, we determine the three-loop symbol uniquely. Both results pass soft, double-soft and directional dual conformal invariance (DDCI) checks, provide the first multi-loop non-MHV form-factor data, and support the universality of the 88-letter alphabet for four-point form factors beyond the MHV sector.

hep-th↗

arXiv2026

QCD Scattering Amplitudes and Prescriptive Unitarity

We present a systematic framework for the maximally-transcendental part of planar QCD scattering amplitudes and perform the first bootstrap computation of six-gluon MHV amplitudes in massless QCD at the symbol level. By analyzing the maximal weight projection of amplitudes at the integrand level, we relate their maximally-transcendental parts to prescriptive unitarity integrals. This reveals a novel analytic structure: the prefactors multiplying the functions of maximal transcendentality are identified with the four-dimensional leading singularities of the theory. As a consequence, these prefactors admit a complete classification and can be computed using on-shell diagrams, a formalism originally developed in $\mathcal{N}{=}4$ super Yang-Mills theory. As a concrete application, we determine the two-loop prefactors for planar MHV gluon amplitudes at arbitrary multiplicity. Combining these prefactors with recent advances in the planar two-loop six-point function space and explicit six-point prescriptive-unitarity input, we construct a complete symbol ansatz and uniquely fix the maximally-transcendental part of the two-loop six-gluon MHV QCD amplitudes by imposing physical constraints. The resulting symbols are expressible in a reduced 137-letter alphabet, suggesting that this alphabet is complete for two-loop six-point massless MHV scattering. We also discuss the implications for multi-collinear splitting and multi-soft functions.

hep-th↗

arXiv2025

Notes on off-shell conformal integrals and correlation functions at five points

We study five-point off-shell conformal integrals and the associated half-BPS correlation functions at two loops in the 't Hooft coupling expansion of maximally supersymmetric Yang-Mills theory. We construct a basis of uniform-transcendental (UT) pure integrals spanning six distinct topologies by diagonalizing leading singularities subject to conformal invariance. By fixing conformal frames, this basis can be mapped to known two-loop four-mass integral families. We then compute the integrated results by combining canonical differential equations with integration-by-parts reduction. As an application, we present symbol-level integrated results for the two-loop five-point half-BPS correlators, including both maximal and non-maximal sectors.

hep-th↗

arXiv2025

Bootstrapping Six-Gluon QCD Amplitudes

We present a symbol-level bootstrap construction of the planar, two-loop six-gluon scattering amplitude for the --++++ helicity configuration in QCD, focusing on the maximal weight pieces-the "most complicated terms" in the sense of Lipatov et al. Building on recent advances in the understanding of the relevant function space, we incorporate as a crucial new ingredient the complete set of leading singularities, obtained from an explicit analysis of on-shell diagrams. The resulting expressions are manifestly conformally invariant and clarify the structure of previous five-particle results. Combining this with the symbol bootstrap, we show that constraints from physical limits are sufficient to uniquely determine the answer. We thus obtain the first concrete characterization of two-loop six-gluon amplitudes at the symbol level and at highest weight. Remarkably, we find that the effective function space involves only 137 symbol letters, significantly fewer than the full set of 167 possible letters, suggesting a yet-unexplained underlying structure akin to that seen in maximally supersymmetric Yang-Mills theory. From the novel amplitude results we extract previously unknown symbol-level results describing two-loop triple collinear and double soft limits.

hep-th↗

arXiv2025

Mastering Cosmological Amplitudes Using Generalized Ramanujan's Theorem

We present a systematic method for computing cosmological amplitudes, including in-in correlators and wavefunction coefficients, in FRW spacetime. Specializing to cases with conformally-coupled external scalars and massive scalar exchanges, we introduce a decomposition into massive family trees, which capture the nested time structure common to these observables. We then evaluate these building blocks using the Method of Brackets (MoB), a multivariate extension of Ramanujan's master theorem that operates directly on the integrand, translating integrals into discrete summations via a compact set of algebraic rules. This yields infinite series representations valid across the full space of external momenta and internal energies. We also develop Feynman-like diagrammatic rules that map interaction graphs to summand structures, enabling efficient and scalable computation. The resulting expressions make time evolution manifest, smoothly interpolate to the conformal limit, and are well suited for both numerical evaluation and analytic analysis of massive field effects in cosmology.

hep-th↗

arXiv2025

Geometric Landau Analysis and Symbol Bootstrap

We investigate how the positive geometry framework for loop integrands in $\mathcal{N}{=}4$ super Yang-Mills theory constrains the structure of the integrated answers. This is done in the context of a geometric expansion of Wilson loops with a Lagrangian insertion, called negative geometries, extending ideas previously used for scattering amplitudes related to the Amplituhedron. The procedure we adopt combines the knowledge of all maximal codimension boundaries of the geometry, which characterize all possible leading singularities of the integral, with a geometrically informed Landau analysis. The interplay between geometry and Landau analysis arises from associating Landau diagrams to geometric boundaries. The boundary structure of the geometry then determines which solutions to the Landau equations are spurious and which ones are physical, that is, which singularities are actually present in the integral. This method allows us to efficiently determine the symbol alphabet of the associated integral, and serves as a starting point for the symbol bootstrap. We successfully implement this procedure and compute the six-point two-loop and five-point three-loop ladder negative geometries at the symbol level. We also present the conjectural alphabet for ladder negative geometries at two loops for all multiplicities. These are finite integrals that serve as building blocks for the Wilson loop with Lagrangian insertion, and therefore provide insights into the function space of the latter.

hep-th↗

arXiv2025

Towards Depth Foundation Model: Recent Trends in Vision-Based Depth Estimation

Depth estimation is a fundamental task in 3D computer vision, crucial for applications such as 3D reconstruction, free-viewpoint rendering, robotics, autonomous driving, and AR/VR technologies. Traditional methods relying on hardware sensors like LiDAR are often limited by high costs, low resolution, and environmental sensitivity, limiting their applicability in real-world scenarios. Recent advances in vision-based methods offer a promising alternative, yet they face challenges in generalization and stability due to either the low-capacity model architectures or the reliance on domain-specific and small-scale datasets. The emergence of scaling laws and foundation models in other domains has inspired the development of "depth foundation models": deep neural networks trained on large datasets with strong zero-shot generalization capabilities. This paper surveys the evolution of deep learning architectures and paradigms for depth estimation across the monocular, stereo, multi-view, and monocular video settings. We explore the potential of these models to address existing challenges and provide a comprehensive overview of large-scale datasets that can facilitate their development. By identifying key architectures and training strategies, we aim to highlight the path towards robust depth foundation models, offering insights into their future research and applications.

cs.CV↗

arXiv2025

Hexagonal Wilson loop with Lagrangian insertion at two loops in $\mathcal{N}=4$ super Yang-Mills theory

In this work, we compute the two-loop result of the null hexagonal Wilson loop with a Lagrangian insertion in planar, maximally supersymmetric Yang-Mills theory via a bootstrap approach. Normalized by the null polygonal Wilson loop itself, the integrand-level result of this observable corresponds to the logarithm of the six-point three-loop amplitude in this theory, while its integrated result is conjectured to match the maximal transcendental part of the six-point three-loop all-plus amplitude in pure Yang-Mills theory. Our work builds on two recent advances. On the one hand, the set of leading singularities relevant to this observable was recently classified. On the other hand, the relevant space of special functions that may in principle accompany these leading singularities was determined at two loops and for six particles by a dedicated Feynman integral calculation. These two ingredients serve as the foundation of our bootstrap ansatz. We fix all indeterminates in this ansatz by imposing physical constraints, such as symmetries, absence of spurious divergences, and correct behavior in soft and collinear limits. Finally, we discuss and verify certain physical properties of our symbol result, including physical singularities, behavior under multi-Regge limit, as well as Steinmann relations between symbol entries. The latter relations are motivated by the correspondence to all-plus amplitudes in pure Yang-Mills theory, and successfully checking them constitutes a consistency check of this conjectured correspondence.

hep-th↗

arXiv2024

Landau-based Schubert analysis

We revisit the conjectural method called Schubert analysis for generating the alphabet of symbol letters for Feynman integrals, which was based on geometries of intersecting lines associated with corresponding cut diagrams. We explain the effectiveness of this somewhat mysterious method by relating such geometries to the corresponding Landau singularities, which also amounts to ``uplifting" Landau singularities of a Feynman integral to its symbol letters. We illustrate this {\it Landau-based Schubert analysis} using various multi-loop Feynman integrals in four dimensions and present an automated {\ttfamily Mathematica} notebook for it. We then apply the method to a simplified problem of studying alphabets of physical quantities such as scattering amplitudes and form factors in planar ${\cal N}=4$ super-Yang-Mills. By focusing on a small set of Landau diagrams (as opposed to all relevant Feynman integrals), we show how this method nicely produces the two-loop alphabet of $n$-point MHV amplitudes and that of the $n=4$ MHV form factors. A byproduct of our analysis is an explicit representation of any symbol alphabet obtained this way as the union of various type-$A$ cluster algebras.

hep-th↗

arXiv2024

Dual conformal invariant kinematics and folding of Grassmannian cluster algebras

Grassmannian manifolds $\Gr(4,n)$ are closely related to the kinematic space of $n$-particle scattering processes in $D=4$, and their combinatorial and geometric structures have played an important role in the study of conformal invariant theories and scattering amplitudes. He, Li, and Yang \cite{HLY26} observed that restricting $D=4$ kinematics to a $D=3$ subspace can be interpreted as a folding of the Grassmannian cluster algebra $\CC[\Gr(4,n)]$ for $n\leq 8$. In this paper, we derive general expressions for the $D=3$ kinematic constraints in terms of Pl\"ucker coordinates of $\Gr(4,n)$ directly from the three-dimensional kinematic condition. We then construct a family of foldable seeds for $\CC[\Gr(4,n)]$, obtained explicitly from the standard initial seed by mutation, whose folding conditions reproduce these kinematic constraints. This establishes the connection between $D=3$ kinematics and folding of Grassmannian cluster algebras for general $n$.

math-ph↗

arXiv2024

From squared amplitudes to energy correlators

The leading order $N$-point energy correlators of maximally supersymmetric Yang-Mills theory in the limit where the $N$ detectors are collinear can be expressed as an integral of the $1\to N$ splitting function, which is given by the $(N{+}3)$-point squared super-amplitudes at tree level. This provides yet another example that the integrand of certain physical observable -- $N$-point energy correlator -- is computed by the canonical form of a positive geometry -- the (tree-level) "squared amplituhedron". By extracting such squared amplitudes from the $f$-graph construction, we compute the integrand of energy correlators up to $N=11$ and reveal new structures to all $N$; we also show important properties of the integrand such as soft and multi-collinear limits. Finally, we take a first look at integrations by studying possible residues of the integrand: our analysis shows that while this gives prefactors in front of multiple polylogarithm functions of $N=3,4$, the first unknown case of $N=5$ already involves elliptic polylogarithmic functions with many distinct elliptic curves, and more complicated curves and higher-dimensional varieties appear for $N>5$.

hep-th↗

arXiv2024

Differential equations and recursive solutions for cosmological amplitudes

Recently considerable efforts have been devoted to computing cosmological correlators and the corresponding wavefunction coefficients, as well as understanding their analytical structures. In this note, we revisit the computation of these ``cosmological amplitudes" associated with any tree or loop graph for conformal scalars with time-dependent interactions in the power-law FRW universe, directly in terms of iterated time integrals. We start by decomposing any such cosmological amplitude (for loop graph, the ``integrand" prior to loop integrations) as a linear combination of {\it basic time integrals}, one for each {\it directed graph}. We derive remarkably simple first-order differential equations involving such time integrals with edges ``contracted" one at a time, which can be solved recursively and the solution takes the form of Euler-Mellin integrals/generalized hypergeometric functions. By combining such equations, we then derive a complete system of differential equations for all time integrals needed for a given graph. Our method works for any graph: for a tree graph with $n$ nodes, this system can be transformed into the {\it canonical differential equations} of size $4^{n{-}1}$ quivalent to the graphic rules derived recently%so-called ``kinematic flow", and we also derive the system of differential equations for loop integrands {\it e.g.} of all-loop two-site graphs and one-loop $n$-gon graphs. Finally, we show how the differential equations truncate for the de Sitter (dS) case (in a way similar to differential equations for Feynman integrals truncate for integer dimensions), which immediately yields the complete symbol for the dS amplitude with interesting structures {\it e.g.} for $n$-site chains and $n$-gon cases.

hep-th↗

arXiv2024

Efficient Model Compression for Hierarchical Federated Learning

Federated learning (FL), as an emerging collaborative learning paradigm, has garnered significant attention due to its capacity to preserve privacy within distributed learning systems. In these systems, clients collaboratively train a unified neural network model using their local datasets and share model parameters rather than raw data, enhancing privacy. Predominantly, FL systems are designed for mobile and edge computing environments where training typically occurs over wireless networks. Consequently, as model sizes increase, the conventional FL frameworks increasingly consume substantial communication resources. To address this challenge and improve communication efficiency, this paper introduces a novel hierarchical FL framework that integrates the benefits of clustered FL and model compression. We present an adaptive clustering algorithm that identifies a core client and dynamically organizes clients into clusters. Furthermore, to enhance transmission efficiency, each core client implements a local aggregation with compression (LC aggregation) algorithm after collecting compressed models from other clients within the same cluster. Simulation results affirm that our proposed algorithms not only maintain comparable predictive accuracy but also significantly reduce energy consumption relative to existing FL mechanisms.

cs.LG↗

arXiv2023

BlockEmulator: An Emulator Enabling to Test Blockchain Sharding Protocols

Numerous blockchain simulators have been proposed to allow researchers to simulate mainstream blockchains. However, we have not yet found a testbed that enables researchers to develop and evaluate their new consensus algorithms or new protocols for blockchain sharding systems. To fill this gap, we developed BlockEmulator, which is designed as an experimental platform, particularly for emulating blockchain sharding mechanisms. BlockEmulator adopts a lightweight blockchain architecture so developers can only focus on implementing their new protocols or mechanisms. Using layered modules and useful programming interfaces offered by BlockEmulator, researchers can implement a new protocol with minimum effort. Through experiments, we test various functionalities of BlockEmulator in two steps. Firstly, we prove the correctness of the emulation results yielded by BlockEmulator by comparing the theoretical analysis with the observed experiment results. Secondly, other experimental results demonstrate that BlockEmulator can facilitate measuring a series of metrics, including throughput, transaction confirmation latency, cross-shard transaction ratio, the queuing status of transaction pools, workload distribution across blockchain shards, etc. We have made BlockEmulator open-source in Github.

cs.CR↗