joaoaleite/PASTEL

PASTEL (Prompted weAk Supervision wiTh crEdibility signaLs) is a weakly supervised approach that leverages large language models to extract credibility signals from web content, then further combines them to predict content veracity without using any ground truth labels.

19
/ 100
Experimental

No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 3 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

3

Forks

—

Language

Jupyter Notebook

License

GPL-3.0

Category

llm-bias-evaluation

Last pushed

Mar 16, 2025

Commits (30d)

0

GitHub
Llm Bias Evaluation · 23 models

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/joaoaleite/PASTEL"

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

Higher-rated alternatives

google-deepmind/long-form-factuality

Benchmarking long-form factuality in large language models. Original code for our paper...

55

gnai-creator/aletheion-llm-v2

Decoder-only LLM with integrated epistemic tomography. Knows what it doesn't know.

38

sandylaker/ib-edl

Calibrating LLMs with Information-Theoretic Evidential Deep Learning (ICLR 2025)

37

nightdessert/Retrieval_Head

open-source code for paper: Retrieval Head Mechanistically Explains Long-Context Factuality

33

MLD3/steerability

An open-source evaluation framework for measuring LLM steerability.

33

Explore Transformer Models

All categories Trending Transformer directory Insights