Systems & Networks · Computer Science & AI
Performance
The newest papers in Performance that ReadSearch has read, newest first. Each links to its own page, with its authors, date, where it was published and the original.
158 papers, from 4 September 2026 to 5 November 2026: 158 arXiv preprints.
Get the few papers worth your time in Performance every morning.
Follow Performance and ReadSearch reads its new papers each day, then brings you the few worth reading, each with its key finding in a sentence.
Newest papers, page 6
GEM-KMeans: Memory-Efficient and Accurate Clustering on Massive Scale with GPU Optimization
Memory-efficient scaling on clustering problems without sacrificing statistical accuracy is of central interest for large-scale data analysis and machine learning problems.Just Let Linear States Forget the Distant Past: Prefix Caching via Suffix Replay for Hybrid LLMs
Hybrid LLMs interleave full-attention layers with linear-attention layers to reduce long-context inference cost, but this structure complicates prefix caching.JET: Justification Evaluation in Transformer
JET uses pretrained language and vision-language models to select among a finite set of answers without additional training.Where Activation Sparsity and KV-Cache Sparsity Cross in LLM Decoding
At each step, decoding one sequence with a large language model rereads the projection weights, whose traffic is fixed, and the key-value (KV) cache, whose traffic grows with context.SparSP: Exploiting Communication Sparsity for Sequence-Parallel Video DiTs
Diffusion Transformers have become the dominant architecture for video generation.FA-Bench: A Benchmark for Word-Level and Phone-Level Forced-Alignment and ASR Timestamps Under Clean and Noisy Conditions
Forced alignment aligns speech audio with a text transcript to generate word and phone timestamps.Change the Product, Keep the Parameters: Associative Algebra Layers for Transformers
Fast matrix multiplication algorithms keep the product fixed and search for a cheaper way to evaluate it.Packets, Transactions and Queues: Design Principles for HFT Systems from a Measurement Study of CME Market Data
HFT systems are conventionally built as a single-threaded event loop, on the rule that every thread hop adds latency.The KV Cache Is the New Memory Wall
Autoregressive LLM inference at long context is bounded by memory bandwidth, not arithmetic throughput, and the binding resource shifts from model weights to the Key-Value (KV) cache as sequence…Joule-Profiler: Profiling the Energy Consumption of Build Automation Tools Made Easy
Build pipelines are integral to modern software development, yet their energy footprint remains largely invisible to practitioners.EAServe: Encode-Aware Disaggregated Serving for Multimodal Large Language Models
Disaggregating the two stages, Prefill and Decode, onto separate GPU pools is now a standard optimization for (text-only) LLM serving.Cache-Aware Conv3D Lowering Across Embedded World-Model Decoders
Generative world models can provide visual rollouts for embodied planning, yet their feasibility on edge devices depends not only on the learned model but also on how the execution runtime…Goal-Persistent Coding Agents as Scientific Performance Engineers: A Fixed-Radius Nearest-Neighbor Case Study
Coding agents can pursue persistent objectives across many tool-use turns, but evidence that general-purpose agents can conduct rigorous scientific performance engineering remains limited.Paging the Experts: A Reproducible Characterization of Flash-Backed MoE Inference on iPhone
Sparse activation reduces mixture-of-experts computation without eliminating the need to store all experts.A Rapid Pipeline for Training and Deploying ML Models on WeBe Band
Developing optimized machine-learning algorithms for edge devices with limited computational and memory resources is challenging, time-consuming, and highly dependent on device-specific constraints.Evaluating the Effect of the Order of Optimization Passes in Quantum Circuit Optimization
Quantum circuit optimization is critical for mitigating the noise inherent in current quantum hardware.DanLing NestedTensor: Composable Multi-Ragged Tensors for Deep Learning
Variable-size inputs are common in deep learning, but dense batching allocates a shared envelope and spends computation on padding.Energy-efficient operation of neural operators for virtual sensing
Virtual sensing repeatedly reconstructs physical fields from changing observations, often on a fixed geometry.RAMP: Robust Adaptive Mixed-Precision Quantization for Edge CPU Vision Models
Deploying deep learning models on edge CPUs is bottlenecked by computational and memory constraints.Distributed Service Orchestration in Edge-Cloud Continuum for Digital Healthcare
Today's digital healthcare services rely on various applications and functions that must be continuously accessible.