Picture for Honglak Lee

Honglak Lee

University of Michigan, Ann Arbor

Subtask-Aware Visual Reward Learning from Segmented Demonstrations

Add code
Feb 28, 2025
Viaarxiv icon

KL Penalty Control via Perturbation for Direct Preference Optimization

Add code
Feb 18, 2025
Viaarxiv icon

Clear Preferences Leave Traces: Reference Model-Guided Sampling for Preference Learning

Add code
Jan 25, 2025
Viaarxiv icon

Probing Visual Language Priors in VLMs

Add code
Dec 31, 2024
Figure 1 for Probing Visual Language Priors in VLMs
Figure 2 for Probing Visual Language Priors in VLMs
Figure 3 for Probing Visual Language Priors in VLMs
Figure 4 for Probing Visual Language Priors in VLMs
Viaarxiv icon

Map2Text: New Content Generation from Low-Dimensional Visualizations

Add code
Dec 24, 2024
Viaarxiv icon

EXAONE 3.5: Series of Large Language Models for Real-world Use Cases

Add code
Dec 09, 2024
Figure 1 for EXAONE 3.5: Series of Large Language Models for Real-world Use Cases
Figure 2 for EXAONE 3.5: Series of Large Language Models for Real-world Use Cases
Figure 3 for EXAONE 3.5: Series of Large Language Models for Real-world Use Cases
Figure 4 for EXAONE 3.5: Series of Large Language Models for Real-world Use Cases
Viaarxiv icon

If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs

Add code
Dec 05, 2024
Figure 1 for If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs
Figure 2 for If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs
Figure 3 for If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs
Figure 4 for If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs
Viaarxiv icon

Interactive and Expressive Code-Augmented Planning with Large Language Models

Add code
Nov 21, 2024
Viaarxiv icon

Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents

Add code
Oct 29, 2024
Figure 1 for Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents
Figure 2 for Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents
Figure 3 for Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents
Figure 4 for Auto-Intent: Automated Intent Discovery and Self-Exploration for Large Language Model Web Agents
Viaarxiv icon

EXAONE 3.0 7.8B Instruction Tuned Language Model

Add code
Aug 07, 2024
Figure 1 for EXAONE 3.0 7.8B Instruction Tuned Language Model
Figure 2 for EXAONE 3.0 7.8B Instruction Tuned Language Model
Figure 3 for EXAONE 3.0 7.8B Instruction Tuned Language Model
Figure 4 for EXAONE 3.0 7.8B Instruction Tuned Language Model
Viaarxiv icon