🏢 Fudan University
Agent-R: Training Language Model Agents to Reflect via Iterative Self-Training
·4105 words·20 mins·
loading
·
loading
AI Generated
🤗 Daily Papers
Natural Language Processing
Large Language Models
🏢 Fudan University
Agent-R: A novel self-training framework enables language model agents to learn from errors by dynamically constructing training data that corrects erroneous actions, resulting in significantly improv…
Dolphin: Closed-loop Open-ended Auto-research through Thinking, Practice, and Feedback
·3489 words·17 mins·
loading
·
loading
AI Generated
🤗 Daily Papers
Computer Vision
3D Vision
🏢 Fudan University
DOLPHIN: AI automates scientific research from idea generation to experimental validation.
LiFT: Leveraging Human Feedback for Text-to-Video Model Alignment
·276 words·2 mins·
loading
·
loading
AI Generated
🤗 Daily Papers
Computer Vision
Video Understanding
🏢 Fudan University
LiFT leverages human feedback, including reasoning, to effectively align text-to-video models with human preferences, significantly improving video quality.
BitStack: Fine-Grained Size Control for Compressed Large Language Models in Variable Memory Environments
·6027 words·29 mins·
loading
·
loading
AI Generated
🤗 Daily Papers
Natural Language Processing
Large Language Models
🏢 Fudan University
BitStack: Dynamic LLM sizing for variable memory!