Add library name, pipeline tag, and Github link
#1
by
nielsr
HF Staff
- opened
README.md
CHANGED
|
@@ -1,8 +1,11 @@
|
|
| 1 |
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
datasets:
|
| 4 |
- yolay/RAIF-ComplexInstruction-DeepSeek
|
|
|
|
|
|
|
|
|
|
| 5 |
---
|
|
|
|
| 6 |
This model belongs to the official implementation of the paper "Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models".
|
| 7 |
|
| 8 |
Existing large language models (LLMs) face challenges of following complex instructions, especially when multiple constraints are present and organized in paralleling, chaining, and branching structures. One intuitive solution, namely chain-of-thought (CoT), is expected to universally improve capabilities of LLMs. However, we find that the vanilla CoT exerts a negative impact on performance due to its superficial reasoning pattern of simply paraphrasing the instructions. It fails to peel back the compositions of constraints for identifying their relationship across hierarchies of types and dimensions.
|
|
@@ -45,7 +48,7 @@ The model DeepScaleR-1.5B is our optimized model for its advanced instruction-fo
|
|
| 45 |
| DeepSeek-Qwen7B | SFT | 67.09 | 69.10 | 58.66 | 58.42 | 55.60 | 65.96 | 79.15 | 64.85 (-0.88%) |
|
| 46 |
| DeepSeek-Qwen7B | Ours | 71.35 | 71.40 | 58.67 | 62.04 | 59.65 | 59.38 | 82.00 | 66.35 (+0.62%) |
|
| 47 |
|
| 48 |
-
|
| 49 |
|
| 50 |
🎓 If you find this work useful, please consider the following citation:
|
| 51 |
```
|
|
|
|
| 1 |
---
|
|
|
|
| 2 |
datasets:
|
| 3 |
- yolay/RAIF-ComplexInstruction-DeepSeek
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
library_name: transformers
|
| 6 |
+
pipeline_tag: text-generation
|
| 7 |
---
|
| 8 |
+
|
| 9 |
This model belongs to the official implementation of the paper "Incentivizing Reasoning for Advanced Instruction-Following of Large Language Models".
|
| 10 |
|
| 11 |
Existing large language models (LLMs) face challenges of following complex instructions, especially when multiple constraints are present and organized in paralleling, chaining, and branching structures. One intuitive solution, namely chain-of-thought (CoT), is expected to universally improve capabilities of LLMs. However, we find that the vanilla CoT exerts a negative impact on performance due to its superficial reasoning pattern of simply paraphrasing the instructions. It fails to peel back the compositions of constraints for identifying their relationship across hierarchies of types and dimensions.
|
|
|
|
| 48 |
| DeepSeek-Qwen7B | SFT | 67.09 | 69.10 | 58.66 | 58.42 | 55.60 | 65.96 | 79.15 | 64.85 (-0.88%) |
|
| 49 |
| DeepSeek-Qwen7B | Ours | 71.35 | 71.40 | 58.67 | 62.04 | 59.65 | 59.38 | 82.00 | 66.35 (+0.62%) |
|
| 50 |
|
| 51 |
+
Code: https://github.com/yuleiqin/RAIF
|
| 52 |
|
| 53 |
🎓 If you find this work useful, please consider the following citation:
|
| 54 |
```
|