《深度学习与计算机视觉》配套代码
Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities. ACM Computing Surveys, 2026.
本人多次机器学习与大数据竞赛Top5的经验总结，满满的干货，拿好不谢
FusionBench: A Comprehensive Benchmark/Toolkit of Deep Model Fusion
AdaMerging: Adaptive Model Merging for Multi-Task Learning. ICLR, 2024.
The is the official implementation of ICML 2023 paper "Revisiting Weighted Aggregation in Federated Learning with Neural Networks".
Compound model panel for pi: parallel model calls, then one synthesis response.
Representation Surgery for Multi-Task Model Merging. ICML, 2024.
Combine the results from a panel of models into an enhanced response
Official repository of the "Transformer Fusion with Optimal Transport" paper, published as a conference paper at ICLR 2024.
Pack of LLMs: Model Fusion at Test-Time via Perplexity Optimization
Ultracode making for OpenCode, inspired by Claude Code
Self-hosted Fusion API and OpenRouter Fusion alternative for building reliable multi-model LLM ensembles — fan out one prompt to N models, then synthesize one answer with a judge model. OpenAI-compatible API, CLI, and eval harness.
ScreamingFace is an open toolkit for composing model fusions: several models answering together, graded against a real research benchmark. Every run reproduces from a single line, on your own keys.
Implementation for the paper "Copyright-Protected Language Generation via Adaptive Model Fusion"
SurgeryV2: Bridging the Gap Between Model Merging and Multi-Task Learning with Deep Representation Surgery. Arxiv, 2024.
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.
[EMNLP 2025] To See a World in a Spark of Neuron: Disentangling Multi-task Interference for Training-free Model Merging
To associate your repository with the model-fusion topic, visit your repo's landing page and select "manage topics."