Picture for Bo Pang

Bo Pang

RLHF Workflow: From Reward Modeling to Online RLHF

Add code
May 13, 2024
Viaarxiv icon

Latent Plan Transformer: Planning as Latent Variable Inference

Feb 07, 2024
Viaarxiv icon

Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts

Dec 01, 2023
Viaarxiv icon

Emergence of Abstract State Representations in Embodied Sequence Modeling

Add code
Nov 07, 2023
Viaarxiv icon

Non-Intrusive Adaptation: Input-Centric Parameter-efficient Fine-Tuning for Versatile Multimodal Modeling

Oct 18, 2023
Figure 1 for Non-Intrusive Adaptation: Input-Centric Parameter-efficient Fine-Tuning for Versatile Multimodal Modeling
Figure 2 for Non-Intrusive Adaptation: Input-Centric Parameter-efficient Fine-Tuning for Versatile Multimodal Modeling
Figure 3 for Non-Intrusive Adaptation: Input-Centric Parameter-efficient Fine-Tuning for Versatile Multimodal Modeling
Figure 4 for Non-Intrusive Adaptation: Input-Centric Parameter-efficient Fine-Tuning for Versatile Multimodal Modeling
Viaarxiv icon

Learning the Geodesic Embedding with Graph Neural Networks

Add code
Sep 11, 2023
Viaarxiv icon

XGen-7B Technical Report

Add code
Sep 07, 2023
Figure 1 for XGen-7B Technical Report
Figure 2 for XGen-7B Technical Report
Figure 3 for XGen-7B Technical Report
Figure 4 for XGen-7B Technical Report
Viaarxiv icon

Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting

Add code
Jun 09, 2023
Figure 1 for Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting
Figure 2 for Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting
Figure 3 for Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting
Figure 4 for Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting
Viaarxiv icon

Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference

Add code
Jun 01, 2023
Figure 1 for Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference
Figure 2 for Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference
Figure 3 for Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference
Figure 4 for Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference
Viaarxiv icon

PaLI-X: On Scaling up a Multilingual Vision and Language Model

May 29, 2023
Figure 1 for PaLI-X: On Scaling up a Multilingual Vision and Language Model
Figure 2 for PaLI-X: On Scaling up a Multilingual Vision and Language Model
Figure 3 for PaLI-X: On Scaling up a Multilingual Vision and Language Model
Figure 4 for PaLI-X: On Scaling up a Multilingual Vision and Language Model
Viaarxiv icon