Explore documented scientific discoveries, original research and new scientific developments. Search the library by topic, category or date.
arXiv:2611.03727v1 Announce Type: new Abstract: Benchmarks built from model hubs combine interaction data, knowledge graphs (KGs) and resource documentation, but valid files do not ensure that these artifacts describe the same objects.
arXiv MLRead article →arXiv:2611.03724v1 Announce Type: new Abstract: In receiver-directed federated learning, each site draws a trust boundary. For every source and model component it judges, on its own data, whether the transfer helps (admit), harms (reject) or cannot yet be told (abstain). We ask whether this boundary holds over hundreds of rounds.
arXiv MLRead article →arXiv:2611.03675v1 Announce Type: new Abstract: Deep neural networks extract powerful latent representations from images, enabling fast and efficient downstream analysis. However, these latent spaces lack explicit geometric structure, leading to possibly poor interpolation fidelity and misleading extrapolation.
arXiv MLRead article →arXiv:2611.03656v1 Announce Type: new Abstract: Why do some of the world's most powerful Artificial Intelligence (AI) systems still struggle to reason about cause and effect? Causal discovery, the ability to discern the underlying drivers of a system, is fundamental to scientific inquiry and has traditionally relied on structured data and statistical methods.
arXiv MLRead article →arXiv:2611.03655v1 Announce Type: new Abstract: Language models often spread a concept across many internal dimensions, making their features difficult to interpret. Sparse autoencoders (SAEs) recover clearer features after training. We ask whether models can instead learn interpretable output coordinates directly.
arXiv MLRead article →arXiv:2611.03654v1 Announce Type: new Abstract: Cardiac auscultation traditionally detects the presence of a heart murmur, but clinicians further characterize it by timing, shape, pitch, quality, and grade. Most automatic phonocardiogram analysis systems stop at binary detection and do not estimate this richer description.
arXiv MLRead article →arXiv:2611.03653v1 Announce Type: new Abstract: Bubeck and Sellke's universal law of robustness says that fitting $n$ noisy samples below the noise level with a class of $p$ real parameters forces a Lipschitz constant of order $\sqrt{nd/p}$, when the covariate distribution on $\mathbb{R}^d$ is isoperimetric: smooth interpolation costs a factor $d$ of overparametrization.
arXiv MLRead article →arXiv:2611.03652v1 Announce Type: new Abstract: Lifelong (continual) anomaly detection systems must learn a sequence of tasks without forgetting how to recognize anomaly types seen earlier -- a property known as backward transfer. A common way to try to improve backward transfer is to reshape a model's internal representation around learnable concept centroids, typically coupled with a new way of scoring anomalies…
arXiv MLRead article →arXiv:2608.15173v2 Announce Type: replace-cross Abstract: Long-running quantum computations, including error-corrected workloads, demand instruction availability, measurement-dependent control, and calibration updates. We present Time-Deterministic and Instruction-Dynamic Execution, an FPGA quantum-control processor supporting runtime revision at both instruction and parameter levels.
arXiv Computer Science ARRead article →arXiv:2604.14626v3 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) models have become the dominant architecture for large-scale language models, yet on-premises serving remains fundamentally memory-bound as batching turns sparse per-token compute into dense memory activation.
arXiv Computer Science ARRead article →arXiv:2509.26494v5 Announce Type: replace-cross Abstract: Syndrome extraction for a CSS code has depth at least the Tanner-graph degree $\Delta$, attained by edge colouring if gate order is ignored; correctness adds a parity constraint.
arXiv Computer Science ARRead article →arXiv:2610.37466v2 Announce Type: replace Abstract: Physical non-idealities in integrated circuit (IC) design, fabrication, and operation are concealed by conventional hardware abstractions, causing unexpected performance variations in kernels. For large-scale workloads, such as AI training and inference, performance unpredictability can lead to substantial losses.
arXiv Computer Science ARRead article →arXiv:2609.22743v4 Announce Type: replace Abstract: Weak-memory processors rely on ordering instructions for correctness, yet conventional implementations often enforce them more conservatively than the memory model requires. This over-enforcement manifests as drain-induced retirement stalls at ordering instructions and conservative squash/replay of speculative loads, suppressing legal executions and reducing…
arXiv Computer Science ARRead article →arXiv:2608.07226v2 Announce Type: replace Abstract: Practical evidence for multi-node training on desktop-class accelerators remains limited. This report presents a proof-of-concept deployment of distributed NanoChat pretraining across two NVIDIA DGX Spark systems, each with a GB10 Grace Blackwell system-on-chip and 128 GB of unified memory, administered remotely over a Tailscale mesh VPN and connected for training…
arXiv Computer Science ARRead article →arXiv:2605.25522v2 Announce Type: replace Abstract: Graph-based Approximate Nearest Neighbor Search (ANNS) has emerged as the dominant approach for large-scale AI applications, offering superior accuracy-efficiency trade-offs over alternative methods. However, graph-based ANNS is inherently memory-bound due to frequent and irregular graph traversal.
arXiv Computer Science ARRead article →arXiv:2506.23901v3 Announce Type: replace Abstract: Novel compute systems are an emerging research topic, aiming towards building next-generation compute platforms. For these systems to thrive, they need to be provided as research infrastructure to allow acceptance and usage by a large community.
arXiv Computer Science ARRead article →arXiv:2611.05233v1 Announce Type: cross Abstract: Fault-tolerant quantum computers run millions of logical cycles to perform quantum error correction (QEC). As logical operations span entire logical cycles, the logical cycle time (LCT) dictates program runtime. Qubit measurements, critical to extract parity information for QEC, are the slowest operations and account for 43-86% of LCT.
arXiv Computer Science ARRead article →arXiv:2611.05104v1 Announce Type: cross Abstract: LLM inference decode executes fine-grained collectives whose latency directly limits per-user interactivity, yet scale-up fabrics are still built from electronic packet switches designed for training.
arXiv Computer Science ARRead article →arXiv:2611.04740v1 Announce Type: cross Abstract: RISC-V based AI accelerators have become popular and not only for AI and ML, but also as vehicles for exploring HPC workloads more generally. Tenstorrent are one example, and whilst they document their Tensix architecture in detail and develop their software stack openly, running code has historically required a physical card, and their recently released simulator…
arXiv Computer Science ARRead article →arXiv:2611.04273v1 Announce Type: cross Abstract: As deep learning (DL) models become increasingly complex and hardware architectures grow more heterogeneous, DL compilers play a critical role in optimizing computational efficiency and enabling scalable deployment.
arXiv Computer Science ARRead article →arXiv:2611.04253v1 Announce Type: cross Abstract: On-demand model loading dominates response latency for interactive on-device AI, yet unified memory architectures (UMAs), now common on edge devices, should make it cheap by placing weights in memory the GPU already sees. However, frameworks still use the loading flow built for discrete GPUs, leaving significant inefficiencies.
arXiv Computer Science ARRead article →arXiv:2611.04222v1 Announce Type: cross Abstract: Under expert parallelism, mixture-of-experts (MoE) layers pay all-to-all token dispatch, grouped GEMMs, and a combine step in every layer. In decode each rank holds only 1-256 tokens: dispatch latency and weight traffic dominate while tensor-core utilization is inherently low.
arXiv Computer Science ARRead article →arXiv:2611.04804v1 Announce Type: new Abstract: 4D Gaussian Splatting (4DGS) enables high-quality real-time novel-view synthesis for dynamic scenes, supporting free-viewpoint video, mobile XR, and autonomous systems at the edge. Methods using explicit 4D distributions, keyframe interpolation, or anchor binding share two GPU data-movement bottlenecks.
arXiv Computer Science ARRead article →arXiv:2611.04144v1 Announce Type: new Abstract: LLMs have rapidly expanded from simple chatbots to agentic AI, where models directly use tools to solve complex problems. To improve problem solving, most frontier LLMs adopt an MoE architecture, pushing model sizes into the trillion-parameter regime.
arXiv Computer Science ARRead article →