Picture for Honglak Lee

Honglak Lee

University of Michigan, Ann Arbor

Deep Exploration of Cross-Lingual Zero-Shot Generalization in Instruction Tuning

Add code
Jun 13, 2024
Viaarxiv icon

MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows

Add code
Jun 10, 2024
Figure 1 for MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows
Figure 2 for MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows
Figure 3 for MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows
Figure 4 for MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows
Viaarxiv icon

Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense

Add code
May 07, 2024
Figure 1 for Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense
Figure 2 for Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense
Figure 3 for Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense
Figure 4 for Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense
Viaarxiv icon

Small Language Models Need Strong Verifiers to Self-Correct Reasoning

Add code
Apr 26, 2024
Figure 1 for Small Language Models Need Strong Verifiers to Self-Correct Reasoning
Figure 2 for Small Language Models Need Strong Verifiers to Self-Correct Reasoning
Figure 3 for Small Language Models Need Strong Verifiers to Self-Correct Reasoning
Figure 4 for Small Language Models Need Strong Verifiers to Self-Correct Reasoning
Viaarxiv icon

Instruction Matters, a Simple yet Effective Task Selection Approach in Instruction Tuning for Specific Tasks

Add code
Apr 25, 2024
Figure 1 for Instruction Matters, a Simple yet Effective Task Selection Approach in Instruction Tuning for Specific Tasks
Figure 2 for Instruction Matters, a Simple yet Effective Task Selection Approach in Instruction Tuning for Specific Tasks
Figure 3 for Instruction Matters, a Simple yet Effective Task Selection Approach in Instruction Tuning for Specific Tasks
Figure 4 for Instruction Matters, a Simple yet Effective Task Selection Approach in Instruction Tuning for Specific Tasks
Viaarxiv icon

Super-resolution of biomedical volumes with 2D supervision

Add code
Apr 15, 2024
Viaarxiv icon

Source-Aware Training Enables Knowledge Attribution in Language Models

Add code
Apr 11, 2024
Viaarxiv icon

View Selection for 3D Captioning via Diffusion Ranking

Add code
Apr 11, 2024
Viaarxiv icon

Step-Calibrated Diffusion for Biomedical Optical Image Restoration

Add code
Mar 26, 2024
Figure 1 for Step-Calibrated Diffusion for Biomedical Optical Image Restoration
Figure 2 for Step-Calibrated Diffusion for Biomedical Optical Image Restoration
Figure 3 for Step-Calibrated Diffusion for Biomedical Optical Image Restoration
Figure 4 for Step-Calibrated Diffusion for Biomedical Optical Image Restoration
Viaarxiv icon

AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents

Add code
Mar 13, 2024
Figure 1 for AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents
Figure 2 for AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents
Figure 3 for AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents
Figure 4 for AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents
Viaarxiv icon