CMU Locus Lab

convex_adversarial

A method for training neural networks that are provably robust to adversarial attacks.

convmixer

Implementation of ConvMixer for "Patches Are All You Need? 🤷"

deq

[NeurIPS'19] Deep Equilibrium Models

mpc.pytorch

A fast and differentiable model predictive control (MPC) solver for PyTorch.

optnet

OptNet: Differentiable Optimization as a Layer in Neural Networks

qpth

A fast and differentiable QP solver for PyTorch.

SATNet

Bridging deep learning and logical reasoning using a differentiable satisfiability solver.

smoothing

Provable adversarial robustness at ImageNet scale

TCN

Sequence modeling benchmarks and temporal convolutional networks

wanda

A simple and effective LLM pruning approach.

locuslab/mpc.pytorch

A fast and differentiable model predictive control (MPC) solver for PyTorch.

locuslab/wanda

A simple and effective LLM pruning approach.

locuslab/qpth

A fast and differentiable QP solver for PyTorch.

locuslab/fast_adversarial

[ICLR 2020] A repository for extremely fast adversarial training using FGSM

locuslab/ect

Consistency Models Made Easy

locuslab/differentiable-mpc

locuslab/open-unlearning

A one-stop repository for large language model (LLM) unlearning. Supports TOFU, MUSE and is an easily extensible framework for new datasets, evaluations, methods, and other benchmarks.

locuslab/massive-activations

Code accompanying the paper "Massive Activations in Large Language Models"

locuslab/cfd-gcn

locuslab/FLYP

Code for Finetune like you pretrain: Improved finetuning of zero-shot vision models

locuslab/torchdeq

Modern Fixed Point Systems using Pytorch

locuslab/projected_sinkhorn

locuslab/edge-of-stability

locuslab/scaling_laws_data_filtering

locuslab/chatllm-vscode

locuslab/deq-ddim

locuslab/lml

The Limited Multi-Label Projection Layer

locuslab/llava-token-compression

locuslab/tta_conjugate

Test-Time Adaptation via Conjugate Pseudo-Labels

locuslab/T-MARS

Code for T-MARS data filtering

locuslab/diffusion-model-hallucination

locuslab/acr-memorization

locuslab/robust_union

[ICML'20] Multi Steepest Descent (MSD) for robustness against the union of multiple perturbation models.

locuslab/llm-idiosyncrasies

Code release for "Idiosyncrasies in Large Language Models"

locuslab/get

Generative Equilibrium Transformer

locuslab/smoothinv

Single Image Backdoor Inversion via Robust Smoothed Classifiers

locuslab/context-parametric-inversion

locuslab/MonotoneDBM

locuslab/tofu

Landing Page for TOFU

locuslab/convcov