Low latency, High Accuracy, Custom Query routers for Humans and Agents. Built by Prithivi Da

Hierarchical RAG architecture scaling to 693K chunks on consumer hardware (4GB VRAM). Features 3-address routing, hybrid vector+graph fusion, and SetFit classification.

Playing with SetFit approach for few-shot transfer for text classification

Slides, Notebook and Data for Presentation: DataHour: Harnessing ML and NLP for Elevated Customer Experiences

Issue report classification demo with SetFit and Ollama for NASA's Flight System software repositories

This repository contains code developed for the SVAR project (Trafikverket 2023-2025).

Hacking SetFit so that it works with integrated gradients.

Repository for the paper "Literary Metaphor Detection with LLM Fine-Tuning and Few-Shot Learning".

SciBERT-based scientific abstract classification using SetFit framework with context injection data augmentation

A multi-head classifier using SetFit for query preprocessing

Weighted Attention Message Pruner (WAMP) Proxy - prunes redundant messages from LLM history to save up to 50% in token costs while preserving core context

Solving the IberAuTexTIfication task from IberLEF 2024

A sentiment analysis API server

Master Thesis: Enhancing Sentiment Analysis: Model Comparison, Domain Adaptation, and Lexicon Evolution in German Data

Toolkit for the paper "Explainable Subjective Stance Classification with SetFit in Political Discourse". The project leverages the SetFit few-shot learning framework, Sentence Transformers architecture, and traditional linguistic ML to enhance explainability in stance classification.

Solving a dataset generated from the OffendES collection

A 6-label toxic comment classifier using SetFit few-shot learning. SHAP explanations, per-label thresholds, uncertainty quantification. PR-AUC 0.95. FastAPI + Docker.