Welcome to the official repository of SINQ! A novel, fast and high-quality quantization method designed to make any Large Language Model smaller while preserving accuracy.
-
Updated
Feb 23, 2026 - Python
Welcome to the official repository of SINQ! A novel, fast and high-quality quantization method designed to make any Large Language Model smaller while preserving accuracy.
This is an official repository for "LAVA: Data Valuation without Pre-Specified Learning Algorithms" (ICLR2023).
Contrastive Explanation (Foil Trees), developed at TNO/Utrecht University
[ICLR24] AutoVP: An Automated Visual Prompting Framework and Benchmark
Official implementation of FedGAT: Generative Autoregressive Transformers for Model-Agnostic Federated MRI Reconstruction (https://arxiv.org/abs/2502.04521)
✨ Official code for our paper: "Uncertainty-o: One Model-agnostic Framework for Unveiling Epistemic Uncertainty in Large Multimodal Models".
Official project website for the AAAI 2022 paper "Stereo Neural Vernier Caliper"
NeurIPS 2025: Graph Your Own Prompt
A model-agnostic library for generating explanations of machine learning predictions, supporting diverse XAI methods like CEM and LIME.
Post-hoc prototype-based explanations with rules for time-series classifiers
Robust regression algorithm that can be used for explaining black box models (Python implementation)
Robust regression algorithm that can be used for explaining black box models (R implementation)
A simple generic (TensorFlow) function that implements the MAML algorithm for regression problems as designed by Chelsea Finn et al. 2017
Codebase for CIKM '24 paper -- PARs: Predicate-based Association Rules for Efficient and Accurate (Model-Agnostic) Anomaly Explanation
Segmented Sampling for Boundary Approximation (SSBA) generates discrete decision boundary points for generating counterfactual explanations or bounded counterfactuals (restricted feature change).
A model-agnostic autonomous reasoning system that improves decision quality by learning from structured failure, strictly separating creativity (LLM) from intelligence (Loop).
Introducing a novel lightweight, post-hoc, single-pass, model-agnostic uncertainty quantification model for pretrained deep neural networks, designed for efficiency, scalability, and compatibility.
RFC analysis agent with empirical validation. Compares custom API tools vs WebSearch-only. Real-world testing proved custom tools produce superior structured output. Model-agnostic MCP server works with Claude, GPT-4, DeepSeek, Gemini.
AI Explainability 360 Toolkit for Time-Series and Industrial Use Cases
agentic CLI tool in rust using openrouter
Add a description, image, and links to the model-agnostic topic page so that developers can more easily learn about it.
To associate your repository with the model-agnostic topic, visit your repo's landing page and select "manage topics."