Arash Bakhtiari
Arash Bakhtiari has 2 papers in our library, in 2024, most of them on Language models & pretraining, Text generation; often with Ammar Ahmad Awan, Ahmed Hassan Awadallah, Alon Benhaim.
No OpenAlex profile matched yet.
📄 2 works in our library · 🔗 Cited 1,042 times in our library · ⭐ Paper Stars ranking: #2360
Fields, topics and concepts
📚 Shelves 7The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 2 worksComputer Science · 2 worksArtificial Intelligence · 2 worksLanguage and Linguistics · 2 worksComputer Vision and Pattern Recognition · 1 workMathematics · 1 workHardware and Architecture · 1 work
🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 2 worksText generation · 1 work
Works by year
2020s
Papers
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone (2024 · 1,134 citations)
The report asks whether a language model small enough for a phone can compete with much larger systems. The authors argue that data quality, not only scale, drives performance. They train phi-3-mini, a 3.8B-parameter decoder, on 3.3T tokens. The data mix combines aggressively filtered public web text with… - DeepSpeed-FastGen: High-throughput Text Generation for LLMs via MII and DeepSpeed-Inference (2024 · 78 citations)
LLM serving systems must deliver high throughput and low latency together, which is hard when prompts are long.
Co-authors here
- Ammar Ahmad Awan ×2
- Ahmed Hassan Awadallah
- Alon Benhaim
- Hany Awadalla
- Harkirat Behl
- Jeff Rasley
- Jianmin Bao
- Jyoti Aneja
- Marah Abdin
- Masahiro Tanaka
- Nguyen Bach
- Reza Yazdani Aminabadi
Cited by
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/authors/arash-bakhtiari/