arXiv subjects

Yunuo Cen

Publications and source records attributed to Yunuo Cen.

9 recordsLinked to original sources

arXiv2026

Continuous Optimization for Satisfiability Modulo Theories on Linear Real Arithmetic

Efficient solutions for satisfiability modulo theories (SMT) are integral in industrial applications such as hardware verification and design automation. Existing approaches are predominantly based on conflict-driven clause learning, which is structurally difficult to parallelize and therefore scales poorly. In this work, we introduce FourierSMT as a scalable and highly parallelizable continuous-variable optimization framework for SMT. We generalize the Walsh-Fourier expansion (WFE), called extended WFE (xWFE), from the Boolean domain to a mixed Boolean-real domain, which allows the use of gradient methods for SMT. This addresses the challenge of finding satisfying variable assignments to high-arity constraints by local updates of discrete variables. To reduce the evaluation complexity of xWFE, we present the extended binary decision diagram (xBDD) and map the constraints from xWFE to xBDDs. We then show that sampling the circuit-output probability (COP) of xBDDs under randomized rounding is equivalent to the expectation value of the xWFEs. This allows for efficient computation of the constraints. We show that the reduced problem is guaranteed to converge and preserves satisfiability, ensuring the soundness of the solutions. The framework is benchmarked for large-scale scheduling and placement problems with up to 10,000 variables and 700,000 constraints, achieving 8-fold speedups compared to state-of-the-art SMT solvers. These results pave the way for GPU-based optimization of SMTs with continuous systems.

cs.AI↗

arXiv2026

DART-ing Through the Drift: Dynamic Tracing of Knowledge Neurons for Adaptive Inference-Time Pruning

Large Language Models (LLMs) exhibit substantial parameter redundancy, particularly in Feed-Forward Networks (FFNs). Existing pruning methods suffer from two primary limitations. First, reliance on dataset-specific calibration introduces significant data dependency and computational overhead. Second, being predominantly static, they fail to account for the evolving subset of knowledge neurons in LLMs during autoregressive generation as the context evolves. To address this, we introduce DART, i.e., Dynamic Attention-Guided Runtime Tracing), a lightweight, training-free method that performs on-the-fly context-based pruning. DART monitors shifts in attention score distributions to infer context changes, dynamically updating neuron-level masks to retain salient parameters. Across ten benchmarks, DART outperforms prior dynamic baseline, achieving accuracy gains of up to 14.5% on LLAMA-3.1-8B at 70% FFN sparsity. Furthermore, DART achieves up to 3x better ROUGE-L scores with respect to static-masked pruning on summarization tasks, with its performance comparable to the original dense models. We conclusively demonstrate that the proposed framework effectively adapts to diverse semantic contexts, preserves model capabilities across both general and domain-specific tasks while running at less than 10MBs of memory for LLAMA-3.1-8B(16GBs) with 0.1% FLOPs overhead. The code is available at https://github.com/seeder-research/DART.

cs.CL↗

arXiv2025

FourierCSP: Differentiable Constraint Satisfaction Problem Solving by Walsh-Fourier Expansion

The Constraint-satisfaction problem (CSP) is fundamental in mathematics, physics, and theoretical computer science. Continuous local search (CLS) solvers, as recent advancements, can achieve highly competitive results on certain classes of Boolean satisfiability (SAT) problems. Motivated by these advances, we extend the CLS framework from Boolean SAT to general CSP with finite-domain variables and expressive constraint formulations. We present FourierCSP, a continuous optimization framework that generalizes the Walsh-Fourier transform to CSP, allowing for transforming versatile constraints to compact multilinear polynomials, thereby avoiding the need for auxiliary variables and memory-intensive encodings. We employ projected subgradient and mirror descent algorithms with provable convergence guarantees, and further combine them to accelerate gradient-based optimization. Empirical results on benchmark suites demonstrate that FourierCSP is scalable and competitive, significantly broadening the class of problems that can be efficiently solved by differentiable CLS techniques and paving the way toward end-to-end neurosymbolic integration.

cs.AI↗

arXiv2024

Analysis of Higher-Order Ising Hamiltonians

It is challenging to scale Ising machines for industrial-level problems due to algorithm or hardware limitations. Although higher-order Ising models provide a more compact encoding, they are, however, hard to physically implement. This work proposes a theoretical framework of a higher-order Ising simulator, IsingSim. The Ising spins and gradients in IsingSim are decoupled and self-customizable. We significantly accelerate the simulation speed via a bidirectional approach for differentiating the hyperedge functions. Our proof-of-concept implementation verifies the theoretical framework by simulating the Ising spins with exact and approximate gradients. Experiment results show that our novel framework can be a useful tool for providing design guidelines for higher-order Ising machines.

cs.AI↗

arXiv2023

Massively Parallel Continuous Local Search for Hybrid SAT Solving on GPUs

Although state-of-the-art (SOTA) SAT solvers based on conflict-driven clause learning (CDCL) have achieved remarkable engineering success, their sequential nature limits the parallelism that may be extracted for acceleration on platforms such as the graphics processing unit (GPU). In this work, we propose FastFourierSAT, a highly parallel hybrid SAT solver based on gradient-driven continuous local search (CLS). This is realized by a novel parallel algorithm inspired by the Fast Fourier Transform (FFT)-based convolution for computing the elementary symmetric polynomials (ESPs), which is the major computational task in previous CLS methods. The complexity of our algorithm matches the best previous result. Furthermore, the substantial parallelism inherent in our algorithm can leverage the GPU for acceleration, demonstrating significant improvement over the previous CLS approaches. We also propose to incorporate the restart heuristics in CLS to improve search efficiency. We compare our approach with the SOTA parallel SAT solvers on several benchmarks. Our results show that FastFourierSAT computes the gradient 100+ times faster than previous prototypes implemented on CPU. Moreover, FastFourierSAT solves most instances and demonstrates promising performance on larger-size instances.

cs.AI↗

arXiv2023

Energy-efficient superparamagnetic Ising machine and its application to traveling salesman problems

The growth of artificial intelligence and IoT has created a significant computational load for solving non-deterministic polynomial-time (NP)-hard problems, which are difficult to solve using conventional computers. The Ising computer, based on the Ising model and annealing process, has been highly sought for finding approximate solutions to NP-hard problems by observing the convergence of dynamic spin states. However, it faces several challenges, including high power consumption due to artificial spins and randomness emulated by complex circuits, as well as low scalability caused by the rapidly growing connectivity when considering large-scale problems. Here, we present an experimental Ising annealing computer based on superparamagnetic tunnel junctions (SMTJs) with all-to-all connections, which successfully solves a 70-city travelling salesman problem (4761-node Ising problem). By taking advantage of the intrinsic randomness of SMTJs, implementing a proper global annealing scheme, and using an efficient algorithm, our SMTJ-based Ising annealer shows superior performance in terms of power consumption and energy efficiency compared to other Ising schemes. Additionally, our approach provides a promising way to solve complex problems with limited hardware resources. Moreover, we propose a crossbar array architecture for scalable integration using conventional magnetic random access memories. Our results demonstrate that the SMTJ-based Ising annealing computer with high energy efficiency, speed, and scalability is a strong candidate for future unconventional computing schemes.

cs.ET↗

arXiv2023

Bilayer-skyrmion-based design of neuron and synapse for spiking neural network

Magnetic skyrmion technology is promising for the next-generation spintronics-based memory and neuromorphic computing due to their small size, non-volatility and low depinning current density. However, the Magnus force originating from the skyrmion Hall effect causes the skyrmion to move along a curved trajectory, which may lead to the annihilation of the skyrmion in a nanotrack during current-induced skyrmion motion. Consequently, circuits utilizing skyrmionic motion need to be designed to limit the impact of the skyrmion Hall effect. In this work, we propose a design of an artificial neuron, and a synapse using the bilayer device consisting of two antiferromagnetically exchange coupled ferromagnetic layers, which achieves robustness against the skyrmion Hall effect by nullifying the Magnus force. Using micromagnetic simulations, we show that the bilayer device can work as an artificial neuron and also as a synapse by modifying its uniaxial anisotropy. We also demonstrate that our proposed skyrmionic synapse has an intrinsic property of perfectly linear and symmetric weight update, which is highly desirable for the synapse operation. A spiking neural network implemented using our proposed synapse and neuron was simulated and showed to achieve 96.23\% accuracy in performing classification on the MNIST handwritten digit dataset.

cond-mat.mes-hall↗

arXiv2022

Design of Spintronics-based Neuronal and Synaptic Devices for Spiking Neural Network Circuits

Topologically stable magnetic skyrmion has a much lower depinning current density that may be useful for memory as well as neuromorphic computing. However, skyrmion-based devices suffer from the Magnus force originating from the skyrmion Hall effect, which may result in unwanted skyrmion annihilation if the magnitude of the driving current gets too large. A design of an artificial neuron and a synapse using a synthetic antiferromagnetically coupled bilayer device, which nullifies the Magnus force, is demonstrated in this work. The leak term in the artificial leaky integrate-and-fire neuron is achieved by engineering the uniaxial anisotropy profile of the neuronal device. The synaptic device has a similar structure as the neuronal device but has a constant uniaxial anisotropy. The synaptic device also has a linear and symmetric weight update, which is a highly desirable trait of an artificial synapse. Neuronal and synaptic devices based on magnetic domain-wall (DW) motion are also studied and compared to skyrmionic devices. Our simulation results show the energy required to perform such operation in DW or skyrmion-based devices is on the order of a few fJ.

cond-mat.mes-hall↗

arXiv2022

CITS: Coherent Ising Tree Search Algorithm Towards Solving Combinatorial Optimization Problems

Simulated annealing (SA) attracts more attention among classical heuristic algorithms because the solution of the combinatorial optimization problem can be naturally mapped to the ground state of the Ising Hamiltonian. However, in practical implementation, the annealing process cannot be arbitrarily slow and hence, it may deviate from the expected stationary Boltzmann distribution and become trapped in a local energy minimum. To overcome this problem, this paper proposes a heuristic search algorithm by expanding search space from a Markov chain to a recursive depth limited tree based on SA, where the parent and child nodes represent the current and future spin states. At each iteration, the algorithm will select the best near-optimal solution within the feasible search space by exploring along the tree in the sense of `look ahead'. Furthermore, motivated by coherent Ising machine (CIM), we relax the discrete representation of spin states to continuous representation with a regularization term and utilize the reduced dynamics of the oscillators to explore the surrounding neighborhood of the selected tree nodes. We tested our algorithm on a representative NP-hard problem (MAX-CUT) to illustrate the effectiveness of this algorithm compared to semi-definite programming (SDP), SA, and simulated CIM. Our results show that above the primal heuristics SA and CIM, our high-level tree search strategy is able to provide solutions within fewer epochs for Ising formulated NP-optimization problems.

cs.AI↗