UQLM: Uncertainty Quantification for Language Models, is a Python package for UQ-based LLM hallucination detection

👽 Out-of-Distribution Detection with PyTorch

📈 SiRE (Simulation-Informed Revenue Extrapolation with Confidence Estimate for Scaleup Companies Using Scarce Time-Series Data), accepted by CIKM'2022 🗽

PyTorch implementation of our ECCV 2022 paper "Rethinking Confidence Calibration for Failure Prediction"

Learning from scratch a confidence measure

Official pytorch implementation of the paper [Adaptive confidence thresholding for monocular depth estimation]

Confidence Estimation for Black Box Automatic Speech Recognition Systems Using Lattice Recurrent Neural Networks https://arxiv.org/abs/1910.11933 or https://ieeexplore.ieee.org/document/9053264

Demo code for GACE: Geometry Aware Confidence Enhancement

[ACL 2025] Revisiting Epistemic Markers in Confidence Estimation: Can Markers Accurately Reflect Large Language Models' Uncertainty?.

Benchmark for "Offline Policy Comparison with Confidence"

KBS 2024 Paper, A Confidence-based Knowledge Integration Framework for Cross-Domain Table Question Answering

Project of ACL 2025 MlingConf: A Comprehensive Investigation of Multilingual Confidence Estimation for Large Language Models

In recent years, the ability to assess the uncertainty of depth estimates in the context of dense stereo matching has received increased attention due to its potential to detect erroneous estimates. Especially, the introduction of deep learning approaches greatly improved general performance, with feature extraction from multiple modalities prov…

Code for "Confidence-Driven Hierarchical Classification of Cultivated Plant Stresses"

Computation of Reliability Statistics: Reliability, Confidence, Assurance

Source code for predicting confidence scores for the samples in t-sne embeddings.

Simple evaluation of classification confidence intervals.

Using Source-Side Confidence Estimation for Reliable Translation into Unfamiliar Languages

VR based simulation platform for soft skills training, integrating AI-driven feedback from the Google Gemini API to track and enhance user performance. It provides real-time, personalized insights to improve communication and professional skills.