The paper introduces a fine‑grain GPU implementation of the partition phase of the Generalized Partition Crossover (GPX) for large‑scale Traveling Salesman Problem (TSP) instances. By reformulating GPX partitioning as a graph‑parallel problem with coalesced memory layouts, ghost‑node transformations, and connected‑component analysis, the authors parallelize key operations such as union of parent tours, splitting of degree‑four vertices, deletion of common edges, and component identification using CUDA. Experiments on instances from 10,000 to 2 million cities show speedups between 48× and 625× over a naive sequential CPU implementation while significantly reducing memory overhead.
arXiv:2607. 07862v1 Announce Type: cross Abstract: The evolution of compute infrastructure has transformed multi-GPU systems into tightly integrated shared-memory structures.
arXiv:2604.18616v2 Announce Type: replace-cross Abstract: LLM coding agents can generate correct GPU kernels, but their performance still trails expert libraries. Reaching peak throughput requires co...
arXiv:2606. 06641v1 Announce Type: new Abstract: We present Accelerated Fourier SAT (AFSAT), a GPU-accelerated solver for pseudo-Boolean satisfiability based on continuous local search (CLS).
arXiv:2609.36303v1 Announce Type: cross Abstract: Recent advances in agentic heuristic design use AI agents and execution feedback to automate algorithm discovery for challenging optimization problem...
arXiv:2607. 05240v1 Announce Type: cross Abstract: Computing-in-Memory (CIM) accelerators execute Matrix-Vector Multiplications (MVMs) in memory, making them a compelling solution for Machine Learning (ML) workloads.
arXiv:2604. 23466v2 Announce Type: replace Abstract: NVIDIA's CUDA Tile (CuTile) introduces a Python-based, tile-centric abstraction for GPU kernel development that aims to simplify programming while retaining Tensor Core and Tensor Memory Accelerator (TMA) efficiency on modern GPUs.
arXiv:2607. 19539v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures increase model capacity without proportionally increasing computation cost and have become a key building block for scaling large language models (LLMs) to trillion-parameter regimes.
The paper introduces SPECTRA, a runtime‑reconfigurable tiled architecture designed to accelerate speculative decoding for large language models on edge devices. SPECTRA adapts its compute engine within each tile between systolic GEMM execution and vector‑lane GEMV execution, while dynamically adjusting tile count, kernel partitioning, and communication patterns across tiles. Experiments on a 20‑tile FPGA prototype demonstrate up to a 2.09× speedup from tile‑level reconfiguration and an additional 1.25× improvement from system‑level adaptability compared to fixed designs.
arXiv:2509. 23722v2 Announce Type: replace-cross Abstract: Pipeline parallelism is widely used to train large language models (LLMs).
The paper introduces COMPASS, an algorithm for the Ordered Clustered Traveling Salesman Problem (OCTSP) that combines search with learning-accelerated routing through parallel sub-solvers. COMPASS improves solutions continuously with more compute, exploits clustered structure to achieve exact solutions in time exponential in cluster size, and outperforms existing methods. It works with general distance matrices, not just coordinate inputs, and scales to 100,000 synthetic nodes and 28,500 real e-commerce nodes, representing the largest reported routing solution over asymmetric distances.
The paper introduces a GPU-resident, batched Levenberg–Marquardt solver that efficiently optimizes constants in tree-based genetic programming for symbolic regression. By using reverse-mode automatic differentiation to assemble per-tree Jacobians in a single backward sweep, the solver’s per-iteration cost becomes independent of the number of constants per tree, achieving up to 510,000 trees per second on an NVIDIA A100. Integrated into EvoGP, the solver enables end-to-end search that recovers governing equations on 10 of 18 constructed problems, a significant improvement over stock EvoGP.
arXiv:2609.13585v1 Announce Type: cross Abstract: Communication has become a bottleneck in distributed training and inference of large models. Overlapping communication with computation at the granul...