Alert button
Picture for Rami Al-Rfou

Rami Al-Rfou

Alert button

Let Your Graph Do the Talking: Encoding Structured Data for LLMs

Feb 08, 2024
Bryan Perozzi, Bahare Fatemi, Dustin Zelle, Anton Tsitsulin, Mehran Kazemi, Rami Al-Rfou, Jonathan Halcrow

Viaarxiv icon

MotionLM: Multi-Agent Motion Forecasting as Language Modeling

Sep 28, 2023
Ari Seff, Brian Cera, Dian Chen, Mason Ng, Aurick Zhou, Nigamaa Nayakanti, Khaled S. Refaat, Rami Al-Rfou, Benjamin Sapp

Figure 1 for MotionLM: Multi-Agent Motion Forecasting as Language Modeling
Figure 2 for MotionLM: Multi-Agent Motion Forecasting as Language Modeling
Figure 3 for MotionLM: Multi-Agent Motion Forecasting as Language Modeling
Figure 4 for MotionLM: Multi-Agent Motion Forecasting as Language Modeling
Viaarxiv icon

Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization

Mar 25, 2023
Bashar Al-Rfooh, Gheith Abandah, Rami Al-Rfou

Figure 1 for Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization
Figure 2 for Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization
Figure 3 for Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization
Figure 4 for Fine-Tashkeel: Finetuning Byte-Level Models for Accurate Arabic Text Diacritization
Viaarxiv icon

Wayformer: Motion Forecasting via Simple & Efficient Attention Networks

Jul 12, 2022
Nigamaa Nayakanti, Rami Al-Rfou, Aurick Zhou, Kratarth Goel, Khaled S. Refaat, Benjamin Sapp

Figure 1 for Wayformer: Motion Forecasting via Simple & Efficient Attention Networks
Figure 2 for Wayformer: Motion Forecasting via Simple & Efficient Attention Networks
Figure 3 for Wayformer: Motion Forecasting via Simple & Efficient Attention Networks
Figure 4 for Wayformer: Motion Forecasting via Simple & Efficient Attention Networks
Viaarxiv icon

Narrowing the Coordinate-frame Gap in Behavior Prediction Models: Distillation for Efficient and Accurate Scene-centric Motion Forecasting

Jun 10, 2022
DiJia Su, Bertrand Douillard, Rami Al-Rfou, Cheolho Park, Benjamin Sapp

Figure 1 for Narrowing the Coordinate-frame Gap in Behavior Prediction Models: Distillation for Efficient and Accurate Scene-centric Motion Forecasting
Figure 2 for Narrowing the Coordinate-frame Gap in Behavior Prediction Models: Distillation for Efficient and Accurate Scene-centric Motion Forecasting
Figure 3 for Narrowing the Coordinate-frame Gap in Behavior Prediction Models: Distillation for Efficient and Accurate Scene-centric Motion Forecasting
Figure 4 for Narrowing the Coordinate-frame Gap in Behavior Prediction Models: Distillation for Efficient and Accurate Scene-centric Motion Forecasting
Viaarxiv icon

VN-Transformer: Rotation-Equivariant Attention for Vector Neurons

Jun 08, 2022
Serge Assaad, Carlton Downey, Rami Al-Rfou, Nigamaa Nayakanti, Ben Sapp

Figure 1 for VN-Transformer: Rotation-Equivariant Attention for Vector Neurons
Figure 2 for VN-Transformer: Rotation-Equivariant Attention for Vector Neurons
Figure 3 for VN-Transformer: Rotation-Equivariant Attention for Vector Neurons
Figure 4 for VN-Transformer: Rotation-Equivariant Attention for Vector Neurons
Viaarxiv icon

SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer

Oct 15, 2021
Tu Vu, Brian Lester, Noah Constant, Rami Al-Rfou, Daniel Cer

Figure 1 for SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer
Figure 2 for SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer
Figure 3 for SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer
Figure 4 for SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer
Viaarxiv icon

nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?

Jun 03, 2021
Mihir Kale, Aditya Siddhant, Noah Constant, Melvin Johnson, Rami Al-Rfou, Linting Xue

Figure 1 for nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?
Figure 2 for nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?
Figure 3 for nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?
Figure 4 for nmT5 -- Is parallel data still relevant for pre-training massively multilingual language models?
Viaarxiv icon

ByT5: Towards a token-free future with pre-trained byte-to-byte models

May 28, 2021
Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel

Figure 1 for ByT5: Towards a token-free future with pre-trained byte-to-byte models
Figure 2 for ByT5: Towards a token-free future with pre-trained byte-to-byte models
Figure 3 for ByT5: Towards a token-free future with pre-trained byte-to-byte models
Figure 4 for ByT5: Towards a token-free future with pre-trained byte-to-byte models
Viaarxiv icon