Improving Calibration for Long-Tailed Recognition (CVPR2021)
-
Updated
Nov 10, 2021 - Python
Improving Calibration for Long-Tailed Recognition (CVPR2021)
[ACL 2026 Main] VL-Calibration: Decoupled Confidence Calibration for Large Vision-Language Models Reasoning
PyTorch implementation of our ECCV 2022 paper "Rethinking Confidence Calibration for Failure Prediction"
[ICCV 2025 CVAMD] The official implementation of the paper "Prompt4Trust: A Reinforcement Learning Prompt Augmentation Framework for Clinically-Aligned Confidence Calibration in Multimodal Large Language Models".
[ACL 2025] Revisiting Epistemic Markers in Confidence Estimation: Can Markers Accurately Reflect Large Language Models' Uncertainty?.
[IEEE Trans. Med. Imaging] The official implementation of the paper "Improving Robustness and Reliability in Medical Image Classification with Latent-Guided Diffusion and Nested-Ensembles".
Service to examine data processing pipelines (e.g., machine learning or deep learning pipelines) for uncertainty consistency (calibration), fairness, and other safety-relevant aspects.
Investigation of how noise perturbations impact neural network calibration and generalisation
[MICCAI 2025] The official implementation of the paper "Exposing and Mitigating Calibration Biases and Demographic Unfairness in MLLM Few-Shot In-Context Learning for Medical Image Classification".
Python framework for high quality confidence estimation of deep neural networks, providing methods such as confidence calibration and ordinal ranking
CW-BASS v2: saturation-aware, reliability-gated pseudo-label selection for semi-supervised semantic segmentation under foundation-model (DINOv2) teachers
Adaptive movement rehabilitation with confidence calibration — novel Movement Calibration Gap metric combining real-time pose estimation with metacognitive self-assessment
Single-file Python library for scanned-document extraction: measured page quality drives adaptive preprocessing, pluggable OCR/VLM backends, schema-driven extraction with bbox provenance, calibrated confidence, and an eval harness. Zero required dependencies.
Code for enhancing Conformal Prediction using Temperature Scaling. Explore more of our work at:
Dependency-free decision-support tool for comparing job offers and career paths with relative scoring, rule checks, calibrated uncertainty, and validation sweeps.
Uncertainty & Confidence Management (UCM): A healthcare AI benchmark suite for uncertainty recognition, justification boundaries, confidence calibration, proportionate action, and reassessment.
Multimodal deepfake detection with explainable AI, robustness validation, and calibrated trust scoring for real-world media.
Trust layer for document→JSON extraction & AI agents: calibrated per-field confidence + source grounding + accept/review abstention on any OCR/VLM. Ships VerifyDocBench, a novel grounding-conditioned conformal method, and an MCP server.
TACT: signed, label-free confidence weighting for self-consistency voting — with the thin-window boundary (2.5–7.5% of items) that explains why six other designs died
To associate your repository with the confidence-calibration topic, visit your repo's landing page and select "manage topics."