This model was trained using influence-guided dataset selection, a technique that uses influence scores to identify the most impactful training data for specific concepts.

Model Description

Base Model: distilbert/distilgpt2
Training Concepts: sentiment analysis, NLP, social media analytics, opinion mining, text classification
Training Method: Influence-guided data selection
Compute Budget: 5 steps per condition
Total Datasets: 5

Training Approach

This model was trained using three different data selection strategies to validate the effectiveness of influence-guided training:

Positive Influence: Datasets with high positive influence scores (most aligned with target concepts)
Random Baseline: Randomly sampled datasets
Negative Influence: Datasets with high negative influence scores (least aligned)

Benchmark Results

Condition	Perplexity ↓	Train Loss ↓	Eval Loss ↓
Positive	12.32	2.2668	2.5111
Random	25.22	3.7177	3.2276
Negative	25.66	3.4759	3.2451

Lower is better for all metrics

Training Datasets

The model was trained on datasets selected through influence scoring:

simana/textclassificationMNLI (Influence: 1.910)
OdiaGenAI/sentiment_analysis_hindi (Influence: -1.656)
Metric-AI/ILUR-news-text-classification-corpus-formatted (Influence: 0.871)
princeton-nlp/SWE-bench (Influence: 0.731)
argilla/end2end_textclassification (Influence: -0.634)

Intended Use

This model demonstrates the effectiveness of influence-guided training for:

Concept-specific language modeling
Data-efficient training
Dataset curation research

Limitations

Trained on a limited compute budget for benchmarking purposes
May not generalize well outside the target concepts: sentiment analysis, NLP, social media analytics, opinion mining, text classification
Performance depends on the quality of influence score estimation

Citation

If you use this model or the influence-guided training approach, please cite:

bibtex

@software{influence_guided_training,
  title = {Influence-Guided Dataset Selection for Language Models},
  author = {Dowser},
  year = {2025},
  url = {https://huggingface.co/vstrandmoe/sentiment-analysis-mental-health}
}

Model Card Contact

For questions or feedback, visit Durinn

Generated by Dowser - Dataset discovery and training optimization