Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -9,7 +9,7 @@ pinned: false
|
|
| 9 |
|
| 10 |
- **[2024-11]** 🤯🤯 We introduce **Multimodal SAE**, the first framework designed to interpret learned features in large-scale multimodal models using Sparse Autoencoders. Through our approach, we leverage LLaVA-OneVision-72B to analyze and explain the SAE-derived features of LLaVA-NeXT-LLaMA3-8B. Furthermore, we demonstrate the ability to steer model behavior by clamping specific features to alleviate hallucinations and avoid safety-related issues.
|
| 11 |
|
| 12 |
-
[GitHub](https://github.com/EvolvingLMMs-Lab/multimodal-sae)
|
| 13 |
|
| 14 |
- **[2024-10]** 🔥🔥 We present **`LLaVA-Critic`**, the first open-source large multimodal model as a generalist evaluator for assessing LMM-generated responses across diverse multimodal tasks and scenarios.
|
| 15 |
|
|
|
|
| 9 |
|
| 10 |
- **[2024-11]** 🤯🤯 We introduce **Multimodal SAE**, the first framework designed to interpret learned features in large-scale multimodal models using Sparse Autoencoders. Through our approach, we leverage LLaVA-OneVision-72B to analyze and explain the SAE-derived features of LLaVA-NeXT-LLaMA3-8B. Furthermore, we demonstrate the ability to steer model behavior by clamping specific features to alleviate hallucinations and avoid safety-related issues.
|
| 11 |
|
| 12 |
+
[GitHub](https://github.com/EvolvingLMMs-Lab/multimodal-sae) | [Paper](https://arxiv.org/abs/2411.14982)
|
| 13 |
|
| 14 |
- **[2024-10]** 🔥🔥 We present **`LLaVA-Critic`**, the first open-source large multimodal model as a generalist evaluator for assessing LMM-generated responses across diverse multimodal tasks and scenarios.
|
| 15 |
|