Picture for Julia Kreutzer

Julia Kreutzer

RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs

Add code
Jul 02, 2024
Viaarxiv icon

LLM See, LLM Do: Guiding Data Generation to Target Non-Differentiable Objectives

Add code
Jul 01, 2024
Viaarxiv icon

The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm

Add code
Jun 26, 2024
Viaarxiv icon

Critical Learning Periods: Leveraging Early Training Dynamics for Efficient Data Pruning

Add code
May 29, 2024
Viaarxiv icon

Aya 23: Open Weight Releases to Further Multilingual Progress

Add code
May 23, 2024
Figure 1 for Aya 23: Open Weight Releases to Further Multilingual Progress
Figure 2 for Aya 23: Open Weight Releases to Further Multilingual Progress
Figure 3 for Aya 23: Open Weight Releases to Further Multilingual Progress
Figure 4 for Aya 23: Open Weight Releases to Further Multilingual Progress
Viaarxiv icon

Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs

Add code
Feb 26, 2024
Viaarxiv icon

Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model

Add code
Feb 12, 2024
Viaarxiv icon

Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning

Add code
Feb 09, 2024
Viaarxiv icon

Intriguing Properties of Compression on Multilingual Models

Add code
Nov 04, 2022
Viaarxiv icon

JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT

Add code
Oct 05, 2022
Figure 1 for JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT
Figure 2 for JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT
Figure 3 for JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT
Figure 4 for JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT
Viaarxiv icon