29 16 2

David Faragó

dball

AI & ML interests

None yet

Recent Activity

upvoted a paper 29 days ago

Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks

upvoted a paper 29 days ago

MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning

upvoted a paper 29 days ago

HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in HuggingFace

View all activity

Organizations

dball's activity

commented 2 papers about 2 months ago

A Survey on Large Language Model based Autonomous Agents

Paper • 2308.11432 • Published Aug 22, 2023 • 2 •

A Survey on Large Language Model based Autonomous Agents

Paper • 2308.11432 • Published Aug 22, 2023 • 2 •

New activity in Test-Time-Training/ttt-mlp-1.3b-pile-8k 3 months ago

Code description

#1 opened 3 months ago by

dball

commented 3 papers 5 months ago

Data curation via joint example selection further accelerates multimodal learning

Paper • 2406.17711 • Published Jun 25, 2024 • 3 •

Data curation via joint example selection further accelerates multimodal learning

Paper • 2406.17711 • Published Jun 25, 2024 • 3 •

Self-Taught Evaluators

Paper • 2408.02666 • Published Aug 5, 2024 • 28 •

New activity in PatronusAI/Llama-3-Patronus-Lynx-8B-Instruct-v1.1 6 months ago

Error when deploying an inference endpoint

#1 opened 6 months ago by

dball

commented 10 papers 10 months ago

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Paper • 2404.02258 • Published Apr 2, 2024 • 104 •

Training LLMs over Neurally Compressed Text

Paper • 2404.03626 • Published Apr 4, 2024 • 22 •

ReFT: Representation Finetuning for Language Models

Paper • 2404.03592 • Published Apr 4, 2024 • 92 •

commented 2 papers 11 months ago

Self-Rewarding Language Models

Paper • 2401.10020 • Published Jan 18, 2024 • 147 •

Self-Consuming Generative Models Go MAD

Paper • 2307.01850 • Published Jul 4, 2023 • 1 •

New activity in dball/zephyr-7b-sft-qlora about 1 year ago

Is the drop in many metrics expected? Why do SFT first if it makes the model worse? Why not do DPO directly on the mistral model?

#1 opened about 1 year ago by

dball