SatvikPraveen/Optimal-Demo-Selection-ICL

Implements and benchmarks optimal demonstration selection strategies for In-Context Learning (ICL) using LLMs. Covers IDS, RDES, Influence-based Selection, Se², and TopK+ConE across reasoning and classification tasks, analyzing the impact of example relevance, diversity, and ordering on model performance across multiple architectures.

25
/ 100
Experimental
No Package No Dependents
Maintenance 13 / 25
Adoption 3 / 25
Maturity 9 / 25
Community 0 / 25

How are scores calculated?

Stars

3

Forks

Language

Jupyter Notebook

License

MIT

Last pushed

Mar 06, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/SatvikPraveen/Optimal-Demo-Selection-ICL"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.