Decrypt

OpenAI Says Benchmark Used to Measure AI Coding Skill Is 'Contaminated'—Here's Why

OpenAI wants to retire the leading AI coding benchmark—and the reasons reveal a deeper problem with how the whole industry measures itself.
GitHub

DeepCode: Open Agentic Coding

We evaluate DeepCode on the PaperBench benchmark (released by OpenAI), a rigorous testbed requiring AI agents to independently reproduce 20 ICML 2024 papers from scratch. The benchmark comprises 8,316 ...
IEEE

GenAlg-SCL: Genetic Algorithm-Based SCL Polar Decoding Toward List Size Optimization

Abstract: Thanks to its outstanding performance, the CRC-aided successive cancellation list (CA-SCL) decoding algorithm has gained widespread acceptance for polar codes. However, its complexity is ...
SecurityWeek

VS Code Configs Expose GitHub Codespaces to Attacks

VS Code-integrated configuration files are automatically executed in Codespaces when the user opens a repository or pull request. The automatic execution of VS Code-integrated configuration files when ...
IEEE

Simultaneous Multi-Rate Threshold Optimization of Protograph-Based Raptor-Like LDPC Codes via Genetic Algorithm

Abstract: Protograph-based Raptor-like (PBRL) LDPC codes, adopted in the 5G NR eMBB data channel, support a wide range of code rates by generating incremental redundancy through XOR operations. As the ...
GitHub

[OSDI'25 Artifact] Achieving Low-Latency Graph-Based Vector Search via Aligning Best-First Search Algorithm with SSD

Welcome to the artifact repository of OSDI'25 accepted paper: Achieving Low-Latency Graph-Based Vector Search via Aligning Best-First Search Algorithm with SSD! This repository contains the ...