jiazhengli commited on
Commit
19a512a
1 Parent(s): b94494e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +34 -12
README.md CHANGED
@@ -9,31 +9,54 @@ base_model: meta-llama/Meta-Llama-3-8B
9
  model-index:
10
  - name: sft_trained_woaqa_llama3
11
  results: []
 
 
 
 
 
 
 
 
12
  ---
13
 
14
- <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
- should probably proofread and complete it, then remove this comment. -->
16
 
17
- # sft_trained_woaqa_llama3
18
 
19
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the sft_wo_aqa_llama3 dataset.
20
- It achieves the following results on the evaluation set:
21
- - Loss: 0.8950
22
 
23
- ## Model description
24
-
25
- More information needed
26
 
27
  ## Intended uses & limitations
28
 
29
- More information needed
30
 
31
  ## Training and evaluation data
32
 
33
- More information needed
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
 
35
  ## Training procedure
36
 
 
 
37
  ### Training hyperparameters
38
 
39
  The following hyperparameters were used during training:
@@ -59,7 +82,6 @@ The following hyperparameters were used during training:
59
  | 0.8548 | 3.16 | 500 | 0.8958 |
60
  | 0.8468 | 3.79 | 600 | 0.8952 |
61
 
62
-
63
  ### Framework versions
64
 
65
  - PEFT 0.10.0
 
9
  model-index:
10
  - name: sft_trained_woaqa_llama3
11
  results: []
12
+ datasets:
13
+ - jiazhengli/Synthetic_Rationale
14
+ - jiazhengli/Rationale_MCTS
15
+ language:
16
+ - en
17
+ metrics:
18
+ - accuracy
19
+ - f1
20
  ---
21
 
 
 
22
 
23
+ # Meta-Llama-3-8B-QLoRA-Assessment-Rationale-sft
24
 
25
+ The model trained with w/o private data from the EMNLP 2024 Paper: Calibrating LLMs with Preference Optimization on Thought Trees for Generating Rationale in Science Question Scoring.
 
 
26
 
27
+ - **Paper:** [Calibrating LLMs with Preference Optimization on Thought Trees for Generating Rationale in Science Question Scoring](https://arxiv.org/abs/2406.19949) (EMNLP 2024 Findings)
28
+ - **GitHub Repository:** [Thought Tree Assessment Repository](https://github.com/lijiazheng99/thought_tree_assessment)
 
29
 
30
  ## Intended uses & limitations
31
 
32
+ This model offers a valuable resource for research in explainable AI within educational technology. The model is trained with **noisy** response-level rationales. This makes them **unsuitable** for direct application in high-stakes assessments without additional verification.
33
 
34
  ## Training and evaluation data
35
 
36
+ We trained and evaluated the model on the [Synthetic Rationale data](https://huggingface.co/datasets/jiazhengli/Synthetic_Rationale), which was generated from the [Rationale MCTS data](https://huggingface.co/datasets/jiazhengli/Rationale_MCTS).
37
+
38
+ ## Citation
39
+
40
+ Please cite the following work if you utilize this model:
41
+
42
+ **BibTeX:**
43
+
44
+ ```bibtex
45
+ @misc{li2024calibratingllmspreferenceoptimization,
46
+ title={Calibrating LLMs with Preference Optimization on Thought Trees for Generating Rationale in Science Question Scoring},
47
+ author={Jiazheng Li and Hainiu Xu and Zhaoyue Sun and Yuxiang Zhou and David West and Cesare Aloisi and Yulan He},
48
+ year={2024},
49
+ eprint={2406.19949},
50
+ archivePrefix={arXiv},
51
+ primaryClass={cs.CL},
52
+ url={https://arxiv.org/abs/2406.19949},
53
+ }
54
+ ```
55
 
56
  ## Training procedure
57
 
58
+ Please refer to our [paper](https://arxiv.org/abs/2406.19949).
59
+
60
  ### Training hyperparameters
61
 
62
  The following hyperparameters were used during training:
 
82
  | 0.8548 | 3.16 | 500 | 0.8958 |
83
  | 0.8468 | 3.79 | 600 | 0.8952 |
84
 
 
85
  ### Framework versions
86
 
87
  - PEFT 0.10.0