Netta1994/setfit_baai_wikisum_gpt-4o_improved-cot-instructions_chat_few_shot_remove_final_evalu

SetFit with BAAI/bge-base-en-v1.5

This is a SetFit model that can be used for Text Classification. This SetFit model uses BAAI/bge-base-en-v1.5 as the Sentence Transformer embedding model. A LogisticRegression instance is used for classification.

The model has been trained using an efficient few-shot learning technique that involves:

1. Fine-tuning a Sentence Transformer with contrastive learning.
2. Training a classification head with features from the fine-tuned Sentence Transformer.

Model Details

Model Description

• Model Type: SetFit
• Sentence Transformer body: BAAI/bge-base-en-v1.5
• Classification head: a LogisticRegression instance
• Maximum Sequence Length: 512 tokens
• Number of Classes: 2 classes

Model Sources

• Repository: SetFit on GitHub
• Paper: Efficient Few-Shot Learning Without Prompts
• Blogpost: SetFit: Efficient Few-Shot Learning Without Prompts

Model Labels

Label	Examples
1	"Reasoning:\ncontext grounded - The answer correctly includes Joan Gaspart's presidency resignation due to the team's poor performance in the 2003 season, whichis supported by the document.\nEvaluation:" 'Reasoning:\nwrong name - The name "Father Josh Carrier" does not appear in the document; the correct name is "Father Joseph Carrier."\nEvaluation:' "Reasoning:\nhallucination - The answer is incorrect, and it's contradicted.\nEvaluation:"
0	'Reasoning:\nhallucination - The answer contains information that contradicts what appears in the document.\nEvaluation:' 'Reasoning:\nirrelevant - The answeris not relevant to what is asked.\nEvaluation:' 'Reasoning:\nContradiction - The answer states Manhattan, but the document clearly indicates that Queens is the borough with the highest population of Asian-Americans.\n\nEvaluation:'

Evaluation

Metrics

Label	Accuracy
all	0.8133

Uses

Direct Use for Inference

First install the SetFit library:

pip install setfit

Then you can load this model and run inference.

from setfit import SetFitModel

# Download from the 🤗 Hub
model = SetFitModel.from_pretrained("Netta1994/setfit_baai_wikisum_gpt-4o_improved-cot-instructions_chat_few_shot_remove_final_evalu")
# Run inference
preds = model("Reasoning:
contradiction - The answer contains information that contradicts what appears in the document.
Evaluation:")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Word count	3	34.4637	148

Label	Training Sample Count
0	79
1	100

Training Hyperparameters

• batch_size: (16, 16)
• num_epochs: (1, 1)
• max_steps: -1
• sampling_strategy: oversampling
• num_iterations: 20
• body_learning_rate: (2e-05, 2e-05)
• head_learning_rate: 2e-05
• loss: CosineSimilarityLoss
• distance_metric: cosine_distance
• margin: 0.25
• end_to_end: False
• use_amp: False
• warmup_proportion: 0.1
• l2_weight: 0.01
• seed: 42
• eval_max_steps: -1
• load_best_model_at_end: False

Training Results

Epoch	Step	Training Loss	Validation Loss
0.0022	1	0.2446	-
0.1116	50	0.2299	-
0.2232	100	0.1175	-
0.3348	150	0.0861	-
0.4464	200	0.0436	-
0.5580	250	0.0233	-
0.6696	300	0.0262	-
0.7812	350	0.0146	-
0.8929	400	0.015	-

Framework Versions

• Python: 3.10.14
• SetFit: 1.1.0
• Sentence Transformers: 3.1.1
• Transformers: 4.44.0
• PyTorch: 2.4.0+cu121
• Datasets: 3.0.0
• Tokenizers: 0.19.1

Citation

BibTeX

@article{https://doi.org/10.48550/arxiv.2209.11055,
    doi = {10.48550/ARXIV.2209.11055},
    url = {https://arxiv.org/abs/2209.11055},
    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
    title = {Efficient Few-Shot Learning Without Prompts},
    publisher = {arXiv},
    year = {2022},
    copyright = {Creative Commons Attribution 4.0 International}
}