Update README.md
Browse files
README.md
CHANGED
|
@@ -34,10 +34,10 @@ AdaMLLM represents our latest advancement in building domain-specific foundation
|
|
| 34 |
</p>
|
| 35 |
|
| 36 |
|
| 37 |
-
- [AdaptLLM](https://huggingface.co/papers/2309.09530)
|
| 38 |
We employ rule-based methods to extract tasks from domain-specific corpora, reformatting them into reading comprehension tasks for continued pre-training. Our 7B finance model outperforms domain-specific models of much larger scales, such as BloombergGPT-50B.
|
| 39 |
|
| 40 |
-
- AdaMLLM
|
| 41 |
We extend supervised task synthesis to multimodality, introducing a unified visual instruction synthesizer to extract instruction-response pairs from domain-specific image-caption pairs. Our synthetic tasks outperform those generated by manual rules, GPT-4, and GPT-4V in improving domain-specific performance for MLLMs.
|
| 42 |
|
| 43 |
|
|
|
|
| 34 |
</p>
|
| 35 |
|
| 36 |
|
| 37 |
+
- **[AdaptLLM](https://huggingface.co/papers/2309.09530): Adapt LLM to domains**
|
| 38 |
We employ rule-based methods to extract tasks from domain-specific corpora, reformatting them into reading comprehension tasks for continued pre-training. Our 7B finance model outperforms domain-specific models of much larger scales, such as BloombergGPT-50B.
|
| 39 |
|
| 40 |
+
- **AdaMLLM: Adapt MLLM to domains**
|
| 41 |
We extend supervised task synthesis to multimodality, introducing a unified visual instruction synthesizer to extract instruction-response pairs from domain-specific image-caption pairs. Our synthetic tasks outperform those generated by manual rules, GPT-4, and GPT-4V in improving domain-specific performance for MLLMs.
|
| 42 |
|
| 43 |
|