Official implementation of "Resolving UnderEdit & OverEdit with Iterative & Neighbor-Assisted Model Editing" (EMNLP 2025 Findings).
Official implementation for Zhong & Le et al., GNNs Also Deserve Editing, and They Need It More Than Once. ICML 2024
Official repository for Targeted Unlearning with Single Layer Unlearning Gradient (SLUG), ICML 2025
[ACL 2024] An Easy-to-use Hallucination Detection Framework for LLMs.
OneEdit: A Neural-Symbolic Collaboratively Knowledge Editing System.
Neuron-Level Sequential Editing for Large Language Models, ACL 2025 Main
Federated Learning Project for Machine Learning and Deep Learning course at Politecnico di Torino
Representation Surgery for Multi-Task Model Merging. ICML, 2024.
A Terminal Editor, with model editing, window spliting, and tree-sitter based code highlighting. More features are being developed!
[AAAI 2025 oral] Attribution Analysis Meets Model Editing: Advancing Knowledge Correction in Vision Language Models with VisEdit
[NeurIPS 2025 B & D] UniEdit: A Unified Knowledge Editing Benchmark for Large Language Models
[ACL 2024] Learning to Edit: Aligning LLMs with Knowledge Editing
AdaMerging: Adaptive Model Merging for Multi-Task Learning. ICLR, 2024.
Federated Learning Under the Lens of Model Editing
Official codes for COLING 2024 paper "Robust and Scalable Model Editing for Large Language Models": https://arxiv.org/abs/2403.17431v1
This project explores how to merge locally fine-tuned models in Federated Learning using model-editing techniques like Task Arithmetic. The goal is to integrate decentralized updates into a single global model while preserving quality and minimizing interference, without accessing raw data.and maintain high model quality in decentralized settings.
[NeurIPS 2024] Knowledge Circuits in Pretrained Transformers
[EMNLP 2024] To Forget or Not? Towards Practical Knowledge Unlearning for Large Language Models
Decomposing and Editing Predictions by Modeling Model Computation
An LLM editor capable of simultaneous large-scale knowledge editing.
Add a description, image, and links to the model-editing topic page so that developers can more easily learn about it.
To associate your repository with the model-editing topic, visit your repo's landing page and select "manage topics."