Airs Bench
AIRS-Bench: an AI Research Science benchmark for quantifying the end-to-end AI research abilities of LLM agents
AI Agent PlatformOpen Source AI & Machine LearningOpen Source
About
The AI Research Science Benchmark is an eval that quantifies the autonomous research abilities of LLM agents in the area of machine learning. AIRS-Bench comprises 20 tasks from state-of-the-art machine learning papers spanning diverse domains such as NLP, Code, Math, biochemical modelling and time series forecasting.
Open Source Health
- Stars
- 115
- Forks
- 9
- License
- Not stated
- Last commit
- 5 months ago
Python
Alternatives to Airs Bench
- RPG_KDD2025by FacebookresearchThis repository provides the code for implementing RPG described in our KDD'25 paper "Generating Long Semantic IDs in Parallel for Recommendation".
- Daggerby Daggerthe missing software stack for CI
- Seamless Communicationby FacebookresearchFoundational Models for State-of-the-Art Speech and Text Translation
- OrienterNetby FacebookresearchSource Code for Paper "OrienterNet Visual Localization in 2D Public Maps with Neural Matching"
- Matrixby FacebookresearchMatrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data generation.
Resources & Links
Related Categories
Using Airs Bench?
Track its cost next to the rest of your stack and get a reminder before it renews.
Add to my stackVendor
Facebookresearch
Publisher of Fairseq
