DolbyUUU/Reinforcement-Calibration-SimCSE

Reinforcement Calibration SimCSE, combining contrastive learning, artificial potential fields, perceptual loss, and RLHF to achieve improved Semantic Textual Similarity (STS) embeddings. PyTorch-based implementations of PerceptualBERT and ForceBasedInfoNCE, along with fine-tuning capabilities via RLHF and evaluation using SentEval.

13
/ 100
Experimental

If you work with text data and need to precisely measure how similar the meaning of different sentences is, this tool helps you create highly accurate sentence embeddings. It takes your raw text sentences as input and produces numerical representations that capture their semantic content more effectively. This is for NLP practitioners, data scientists, or researchers who need to quantify sentence meaning for tasks like document clustering, search, or content recommendation.

No commits in the last 6 months.

Use this if you need to generate high-quality numerical representations of sentences for Semantic Textual Similarity (STS) tasks and are looking for methods that incorporate human feedback.

Not ideal if you are looking for a pre-trained, ready-to-use model without any fine-tuning or if your primary need is not focused on semantic similarity.

Natural Language Processing Text Analytics Information Retrieval Semantic Search Content Understanding
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

11

Forks

Language

Python

License

Last pushed

Dec 24, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/DolbyUUU/Reinforcement-Calibration-SimCSE"

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