Loading README.md +0 −4 Original line number Diff line number Diff line Loading @@ -128,10 +128,6 @@ We present several cases selected from validation steps during training models w # Project Roadmap - 🗓️ Mar 25, 2025: We release VAGEN, a multi-turn reinforcement learning framework for training VLM Agents! - [ ] Merge to RAGEN for better package mangement - [ ] Address training stability issues - [ ] Lora Support - [ ] Implement improved early stopping and gradient clipping strategies - [ ] Reward normalization (refer to RAGEN) - [ ] Expand evaluation framework to more diverse visual environments - [ ] Scaling to larger models and applying TRICO to text-only tasks Loading Loading
README.md +0 −4 Original line number Diff line number Diff line Loading @@ -128,10 +128,6 @@ We present several cases selected from validation steps during training models w # Project Roadmap - 🗓️ Mar 25, 2025: We release VAGEN, a multi-turn reinforcement learning framework for training VLM Agents! - [ ] Merge to RAGEN for better package mangement - [ ] Address training stability issues - [ ] Lora Support - [ ] Implement improved early stopping and gradient clipping strategies - [ ] Reward normalization (refer to RAGEN) - [ ] Expand evaluation framework to more diverse visual environments - [ ] Scaling to larger models and applying TRICO to text-only tasks Loading
