alirezamsh (Alireza Mohammadshahi)

upvoted a paper 13 days ago

OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI

Paper • 2406.12753 • Published 21 days ago • 14

upvoted 6 papers 2 months ago

A Careful Examination of Large Language Model Performance on Grade School Arithmetic

Paper • 2405.00332 • Published May 1 • 30

Octopus v4: Graph of language models

Paper • 2404.19296 • Published Apr 30 • 116

Better & Faster Large Language Models via Multi-token Prediction

Paper • 2404.19737 • Published Apr 30 • 69

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

Paper • 2404.18796 • Published Apr 29 • 67

AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs

Paper • 2404.16873 • Published Apr 21 • 27

FlowMind: Automatic Workflow Generation with LLMs

Paper • 2404.13050 • Published Mar 17 • 32

upvoted a collection 2 months ago

OpenMath

Collection

A collection of models and datasets introduced in "OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset" • 15 items • Updated 25 days ago • 32

upvoted a paper 2 months ago

Textbooks Are All You Need II: phi-1.5 technical report

Paper • 2309.05463 • Published Sep 11, 2023 • 84

upvoted 2 articles 2 months ago

Article

Synthetic data: save money, time and carbon with open source

Feb 16

• 35

Article

Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models

Mar 20

• 36

upvoted 2 papers 2 months ago

Insights into Alignment: Evaluating DPO and its Variants Across Multiple Tasks

Paper • 2404.14723 • Published Apr 23 • 10

Layer Skip: Enabling Early Exit Inference and Self-Speculative Decoding

Paper • 2404.16710 • Published Apr 25 • 56

upvoted a paper 3 months ago

OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework

Paper • 2404.14619 • Published Apr 22 • 124

upvoted a collection 3 months ago

Top 10% instruction tuning datasets

Collection

Collects datasets with 'instruction' in the name and more than 1 download and in the top 10% for the number of likes • 13 items • Updated 6 days ago • 6

upvoted a paper 3 months ago

Judging LLM-as-a-judge with MT-Bench and Chatbot Arena

Paper • 2306.05685 • Published Jun 9, 2023 • 26

upvoted an article 3 months ago

Article

Mixture of Depth is Vibe

By

•

Apr 22

• 40

upvoted 3 papers 3 months ago

upvoted an article 3 months ago

Article

Mergoo: Efficiently Build Your Own MoE LLM

By

•

Jun 3

• 36

upvoted 3 papers 3 months ago

Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM

Paper • 2403.07816 • Published Mar 12 • 37

Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences

Paper • 2404.03715 • Published Apr 4 • 58

Stream of Search (SoS): Learning to Search in Language

Paper • 2404.03683 • Published Apr 1 • 21

upvoted an article 3 months ago

Article

Orchestration of Experts: The First-Principle Multi-Model System

By

•

May 30

• 14

upvoted a paper 4 months ago

Orca-Math: Unlocking the potential of SLMs in Grade School Math

Paper • 2402.14830 • Published Feb 16 • 23

upvoted a paper 5 months ago

Leeroo Orchestrator: Elevating LLMs Performance Through Model Integration

Paper • 2401.13979 • Published Jan 25 • 2

upvoted 2 collections 7 months ago

Awesome feedback datasets

Collection

A curated list of datasets with human or AI feedback. Useful for training reward models or applying techniques like DPO. • 19 items • Updated Apr 12 • 58

Awesome SFT datasets

Collection

A curated list of interesting datasets to fine-tune language models with. • 43 items • Updated Apr 12 • 101

upvoted 4 papers 7 months ago

SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages

Paper • 2210.11621 • Published Oct 20, 2022 • 1

RQUGE: Reference-Free Metric for Evaluating Question Generation by Answering the Question

Paper • 2211.01482 • Published Nov 2, 2022 • 1

Investigating Multi-Pivot Ensembling with Massively Multilingual Machine Translation Models

Paper • 2311.07439 • Published Nov 13, 2023 • 1

What Do Compressed Multilingual Machine Translation Models Forget?

Paper • 2205.10828 • Published May 22, 2022 • 1

Alireza Mohammadshahi

AI & ML interests

Articles

Mergoo: Efficiently Build Your Own MoE LLM

Orchestration of Experts: The First-Principle Multi-Model System

Organizations

alirezamsh's activity

OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI

A Careful Examination of Large Language Model Performance on Grade School Arithmetic

Octopus v4: Graph of language models

Better & Faster Large Language Models via Multi-token Prediction

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs

FlowMind: Automatic Workflow Generation with LLMs

OpenMath

Textbooks Are All You Need II: phi-1.5 technical report

Synthetic data: save money, time and carbon with open source

Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models

Insights into Alignment: Evaluating DPO and its Variants Across Multiple Tasks

Layer Skip: Enabling Early Exit Inference and Self-Speculative Decoding

OpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework

Top 10% instruction tuning datasets

Judging LLM-as-a-judge with MT-Bench and Chatbot Arena

Mixture of Depth is Vibe

Learning to Route Among Specialized Experts for Zero-Shot Generalization

Instruction-Following Evaluation for Large Language Models

Dataset Reset Policy Optimization for RLHF

Mergoo: Efficiently Build Your Own MoE LLM

Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM

Direct Nash Optimization: Teaching Language Models to Self-Improve with General Preferences

Stream of Search (SoS): Learning to Search in Language

Orchestration of Experts: The First-Principle Multi-Model System

Orca-Math: Unlocking the potential of SLMs in Grade School Math

Leeroo Orchestrator: Elevating LLMs Performance Through Model Integration

Awesome feedback datasets

Awesome SFT datasets

SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages

RQUGE: Reference-Free Metric for Evaluating Question Generation by Answering the Question

Investigating Multi-Pivot Ensembling with Massively Multilingual Machine Translation Models

What Do Compressed Multilingual Machine Translation Models Forget?