⚙️ Systems Programming - hello · Scour

LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments 🔐Futexes

Sketch-and-Verify: Structured Inference-Time Scaling via Program Sketching ✏️Count-Min Sketch

Beyond Static Policies: Exploring Dynamic Policy Selection for Single-Thread Performance Optimization 📍CPU Pinning

t\"{a}k\={o}Formal: Enabling Robust Software for Programmable Memory Hierarchies (Extended Version) 🚧Memory Barriers

VDCores: Resource Decoupled Programming and Execution for Asynchronous GPU 🧵Lightweight Threads

arxiv.org·6d·Hacker News

CCL-Bench 1.0: A Trace-Based Benchmark for LLM Infrastructure 🚀Performance

arxiv.org·4d·Hacker News

Constraint-Aware Execution Planning for Hybrid Space-Ground Compute Workloads ⏰Timely Dataflow

Interaction Tree Semantics for RISC-V: Bridging Compiler and Hardware Verification 👻Phantom Types

Enhancing Performance Insight at Scale: A Heterogeneous Framework for Exascale Diagnostics 📊Extrae

FTPrimitiveBench: A Benchmark Suite For Logical Computation Under Hardware-Motivated and Biased Noise Models 🕰️Hybrid Logical Clocks

From Agent Loops to Deterministic Graphs: Execution Lineage for Reproducible AI-Native Work ⚓Anchors

FalconGEMM: Surpassing Hardware Peaks with Lower-Complexity Matrix Multiplication ⚡Hardware Acceleration

Mechanical Conscience: A Mathematical Framework for Dependability of Machine Intelligenc ⚓Anchors

LCM: Lossless Context Management 📏Linear Types

arxiv.org·5d·Hacker News

Teaching LLMs Program Semantics via Symbolic Execution Traces 🔍KLEE

KEET: Explaining Performance of GPU Kernels Using LLM Agents 🧩mimalloc

Log in to enable infinite scrolling