LeapLab

UltraStar. [MICCAI 2026] UltraStar: Semantic-Aware Star Graph Modeling for Echocardiography Navigation

AdaptiveNN-Jittor. Python

VA-Adapter. [MICCAI 2026] VA-Adapter: Adapting Ultrasound Foundation Model to Echocardiography Probe Guidance

WeightFormer. Python

RvR. 🔥 Regeneration over editing: unlocking more effective image refinement!

CDLM. Python

ParallelBench. [ICLR 2026] ParallelBench: Understanding the Tradeoffs of Parallel Decoding in Diffusion LLMs

UltraHiT. [ICRA 2026] UltraHiT: A Hierarchical Transformer Architecture for Generalizable Internal Carotid Artery Robotic Ultrasonography

AdaGen. Official code for "AdaGen: Learning Adaptive Policy for Image Synthesis"

JustGRPO. [ICML 2026 Outstanding Paper] Minimalist RL for Diffusion LLMs. 89.1% on GSM8K.

Circulant-Attention. [AAAI 2026] Official repository of Circulant Attention

ViTTT. [CVPR 2026] [Best Paper Finalist] [Oral] Official repository of Vision Test-Time Training

4DLangSplat. Official implementation of “4D LangSplat: 4D Language Gaussian Splatting via Multimodal Large Language Models” (CVPR 2025)

UltraSeP. [Pattern Recognition 2026] UltraSeP: Sequence-aware Pre-training for Echocardiography Probe Guidance

AdaptiveNN. [Nature Machine Intelligence 2025] Emulating Human-like Adaptive Vision for Efficient and Flexible Machine Visual Perception

IMG-Multimodal-Diffusion-Alignment. IMG: Calibrating Diffusion Models via Implicit Multimodal Guidance, ICCV 2025

Interleaving-Reasoning-Generation. [ICLR 2026] This is an early exploration to introduce Interleaving Reasoning to Text-to-image Generation field and achieve the SoTA benchmark performance. It also significantly improves the quality, fine-grained details and aesthetic aspects of generated images.

CheXWorld. [CVPR 2025] CheXWorld: Exploring Image World Modeling for Radiograph Representation Learning

art-msra. [CVPR 2025] Official repo for ART:Anonymous Region Transformer for Variable Multi-Layer Transparent Image Generation

UltraBot. [Nature Communications 2025] Towards Expert-level Autonomous Carotid Ultrasonography with Large-scale Learning-based Robotic System

cooragent. Official Repository of Cooragent.

MMaDA. MMaDA - Open-Sourced Multimodal Large Diffusion Language Models (dLLMs with block diffusion, mixed-CoT, unified RL)

Absolute-Zero-Reasoner. Official Repository of Absolute Zero Reasoner

GridMix. Repository of GridMix (ICLR 2025)

limit-of-RLVR. repo for paper https://arxiv.org/abs/2504.13837

CODA. CODA: Repurposing Continuous VAEs for Discrete Tokenization

DeeR-VLA. Official code of paper "DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution"

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

chameleon. Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.

OVM3D-Det. Python

diver-ct. Python

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

Uni-AdaFocus. Official repository of Uni-AdaFocus (TPAMI 2024).

calvin. CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

InLine. [NeurIPS 2024] Official repository of InLine attention

ENAT. [NeurIPS 2024] ENAT: Rethinking Spatial-temporal Interactions in Token-based Image Synthesis

Attention-Mediators. [ECCV 2024] Efficient Diffusion Transformer with Step-wise Dynamic Attention Mediators

Segment3D. Python

AdaNAT. [ECCV 2024] AdaNAT: Exploring Adaptive Policy for Token-Based Image Generation

mamba-train. A single repo with all scripts and utils to train / fine-tune the Mamba model with or without FIM

UniTTA. Python

ImprovedNAT. A PyTorch implementation of the paper "Revisiting Non-Autoregressive Transformers for Efficient Image Synthesis"

MLLA. [NeurIPS 2024] Official repository of MLLA

ProCo. [TPAMI 2024] Probabilistic Contrastive Learning for Long-Tailed Visual Recognition

SimPro. [ICML 2024] SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning

GSVA. [CVPR2024] GSVA: Generalized Segmentation via Multimodal Large Language Models

ExpeL. Python

LearnableISDA. [IEEE TIP] Fine-grained Recognition with Learnable Semantic Data Augmentation

Smooth-Diffusion. Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models arXiv 2023 / CVPR 2024

Agent-Attention. [ECCV 2024] Official repository of Agent Attention

LLMRiddles. Open-Source Reproduction/Demo of the LLM Riddles Game

FamO2O. Repository of "Train Once, Get a Family: State-Adaptive Balances for Offline-to-Online Reinforcement Learning" (NeurIPS 2023 Spotlight)

Rank-DETR. [NeurIPS 2023] Rank-DETR for High Quality Object Detection

DAT-Jittor. Jittor implementation of Vision Transformer with Deformable Attention

MOSS. Official implementation of A Mixture of Surprises for Unsupervised Reinforcement Learning

L2W-DEN. [ECCV 2022] Learning to Weight Samples for Dynamic Early-exiting Networks

Text4Point.

DAPrompt. Pytorch implementation of DAPrompt: https://arxiv.org/abs/2202.06687

Slide-Transformer. Official repository of Slide-Transformer (CVPR2023)

ActiveNeRF. Official repository of ActiveNeRF (ECCV2022)

Dynamic_Perceiver. Official implementation of Dynamic Perceiver

AdaAFforPINNs. Python

ARC. [ICCV 2023] Adaptive Rotated Convolution for Rotated Object Detection

LASNet. [NeurIPS 2022] Latency-aware Spatial-wise Dynamic Networks

DAT-Detection. Repository of Vision Transformer with Deformable Attention (CVPR2022) and DAT++: Spatially Dynamic Vision Transformerwith Deformable Attention

DAT-Segmentation. Repository of Vision Transformer with Deformable Attention (CVPR2022) and DAT++: Spatially Dynamic Vision Transformerwith Deformable Attention

LAUDNet. [IEEE TPAMI] Latency-aware Unified Dynamic Networks for Efficient Image Recognition

FLatten-Transformer. [ICCV 2023] Official repository of FLatten Transformer

EfficientTrain. 1.5−3.0× lossless training or pre-training speedup. An off-the-shelf, easy-to-implement algorithm for the efficient training of foundation visual backbones.

Deep-Incubation. Code release for Deep Incubation (https://arxiv.org/abs/2212.04129)

Painter. Painter & SegGPT Series: Vision Foundation Models from BAAI

Cross-Modal-Adapter. [Pattern Recognition 2025] Cross-Modal Adapter for Vision-Language Retrieval

CheckpointKD. Python

Pseudo-Q. [CVPR 2022] Pseudo-Q: Generating Pseudo Language Queries for Visual Grounding

DAT. Repository of Vision Transformer with Deformable Attention (CVPR2022) and DAT++: Spatially Dynamic Vision Transformerwith Deformable Attention

AdaFocusV2. [CVPR 2022] Official repository of AdaFocusV2.

ACmix. Official repository of ACmix (CVPR2022)