Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[ICML2026] VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning
| Date | Stars |
|---|---|
| 2026-07-31 | 322 |
| 2026-08-02 | 323 |
| 2026-08-03 | 324 |
| 2026-08-06 | 324 |
| 2026-08-07 | 324 |
| 2026-08-14 | 324 |
| 2026-08-15 | 324 |
| 2026-08-16 | 325 |
| 2026-08-18 | 325 |
| 2026-08-19 | 327 |
| 2026-08-26 | 328 |
| 2026-08-27 | 329 |
| 2026-08-31 | 329 |
| 2026-09-03 | 330 |
| 2026-09-08 | 331 |
| 2026-09-20 | 331 |
Today
— stars today
This week
— stars this week
This month
+4 stars this month
Momentum
0.0
growth rate 0.00%/day
# VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning <div align="center"> [[ICML2026]](https://openreview.net/pdf?id=i7mfaYYLDf) [[paper]](https://arxiv.org/abs/2509.15937) [[code]](https://github.com/InternRobotics/VLAC) [[model-2b]](https://huggingface.co/InternRobotics/VLAC) [[model-8b]](https://huggingface.co/InternRobotics/VLAC-8b) </div> ## VLAC-cut is ready A new Video Critic Model that can cut out bad or good segments in embodied trajectories. A dataset with rich progress annotations that includes both success and failure trajectories. <div align="center"> [[Paper]](https://arxiv.org/abs/2607.09776) [[Code]](https://github.com/InternRobotics/VLAC-cut) [[Model 30B-A3B]](https://huggingface.co/InternRobotics/VLAC-Cut) [[Benchmark]](https://huggingface.co/datasets/InternRobotics/VLAC-Cut-Benchmark) [[Full Data]](https://huggingface.co/datasets/InternRobotics/VLAC-Cut-FullData) </div> ## 🚀 Interactive Demo & Homepage <div align="center"> ### [🎮 **Try Interactive & Homepage**](https://vlac.intern-ai.org.cn/) > **Online Demo is available now in Homepage, Try as you like!!!** </div> <div align="center"> <img src="data/title_banner-2.gif" alt="VLAC banner" width="800"></img> </div> ## VLAC VLAC is a general-purpose pair-wise critic and manipulation model which designed for real world robot reinforcement learning and data refinement. It provides robust evaluation capabilities for task progress prediction and task completion verification base one images and task description. VLAC trained on 3000h+ human egocentric data, 1200h+ comprehensive public robotic manipulation data, and 15h+ self-collected manipulation data. ### VLAC-8 is avaliable now [[model-8b]](https://huggingface.co/InternRobotics/VLAC-8b) ## ✨ Key Features • **Pair-wise comparison mechanism** for improved progressing dense critic accuracy, better recognition of state changes, and each step can be the start of the trajectory. • **Multi-modal capabilities** - Supports process tracking, task completion judgment, task description estimation, visual question answering, and even embodied action output, equipped with VLA capabilities. • **Flexible zero-shot and one-shot** - in-context capabilities, maintaining excellent performance across entities, scenarios, and tasks. • **Human-task synesthesia** - Based on the ego4D human dataset, model understands common tasks and build synesthesia for real-world human tasks and embodied tasks. • **Trajectory quality screening** - VLAC can evaluate the collected trajectories and filters out low score trajectories based on the VOC value and mask the action with negative pair-wise score, that is, data with low fluency and quality, improving the effect and efficiency of imitation learning. ## Framework <div align="center"> <img src="data/framework.png" alt="VLAC Framework" width="800"/> </div> *The VLAC model is trained on a combination of comprehensive public robotic manipulation datasets, human demonstration data, self-collected manipulation data, and various image understanding datasets. Video data is processed into pair-wise samples to learn the different task progress between any two frames, supplemented with task descriptions and task completion evaluation to enable task progress understanding and action generation, as illustrated in the bottom-left corner. As shown in the diagram on the right, the model demonstrates strong generalization capabilities to new robots, scenarios, and tasks not covered in the training dataset. It can predict task progress and distinguish failure action or trajectory, providing dense reward feedback for real-world reinforcement learning and offering guidance for data refinement. Additionally, the model can directly perform manipulation tasks, exhibiting zero-shot capabilities to handle different scenarios.* ## Performance Details about the model's performance and evaluation metrics can be found in the [Homepage](https://vlac.
Excerpt of 8,807 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:102757fe4941acf9, llm:Repository title and description: "VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning" (Python).
matched fp:102757fe4941acf9, llm:Repository title and description: "VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning" (Python).
matched fp:102757fe4941acf9, llm:Repository title and description: "VLAC: A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning" (Python).