model-ensemble

FusionBench: A Comprehensive Benchmark/Toolkit of Deep Model Fusion

This repo contains the code for solving Poisson Equation using Physics Informed Neural Networks

Ensemble based transfer learning approach for accurately classifying common thoracic diseases from Chest X-Rays

This repository serves as a template for creating new projects based on FusionBench. It includes all the necessary configurations and boilerplate code to get started quickly.

Consensus Learning from Heterogeneous Objectives for One-Class Collaborative Filtering (WWW'22)

Biendata astradata competition 1st place solution. (https://www.biendata.com/competition/astrodata2019/)

Implementation of Boosting Certified $\ell_\infty$-dist Robustness with EMA Method and Ensemble Model

To address the impact of rising house prices on the economy, we built a machine learning model resistant to market trends. We experimented with Random Forest and Linear Regression models, employing sophisticated imputation methods like median state price replacement, KNN imputation, and forward/backward filling to minimize errors.

TFPNER: Exploration on the Named Entity Recognition of Token Fused with Part-of-Speech

[網路安全的資料科學 108-2@NCCU] 惡意程式偵測 - 使用靜態分析與模型集成

Amazon ML Challenge 2025: Multimodal Price Prediction Solution

"This repository contains implementations of Boosting method, popular techniques in Model Ensembles, aimed at improving predictive performance by combining multiple models. by using titanic database."

A full-stack machine learning architecture for food delivery ETA prediction, leveraging a DVC-driven pipeline, automated CI/CD workflows, cloud artifact management, and LGBM-based stacked regression ensemble for high-fidelity time estimations.

Multi-model ensemble voting strategies for LLM reliability

Uncertainty-aware deep learning ensemble that turns raw Sentinel-2 imagery into georeferenced land-cover maps with per-pixel confidence - five fused segmentation models, 27 benchmarked experiments, and a decade of measured urban expansion over Vilnius.

Comparison of several machine learning classification models for a fingerprint spoofing task. Developed as academic project at Polytechnic of Turin, A.Y. 2023-24

Selective chest X-ray interpretation that knows when to defer — cross-model VLM disagreement + conformal prediction for trustworthy, domain-agnostic automation.

A systematic study of training recipes, data augmentation, distribution shift, and ensemble strategies for weather image classification.

A diverse 6-LLM judge committee (4 labs, 3 countries) has correlated errors: ~1.9 effective independent judges of 6 (mean pairwise error-correlation ρ≈0.42) on RAGTruth. Provider diversity does not buy error-independence. Verdicts + one-command recompute included.