Collection of awesome parameter-efficient fine-tuning resources.

🌐 Event Camera Vision in the Era of Large Models: A Survey

Signal Propagation Theory for Long-Horizon Agent RL

复现大模型相关算法及一些学习记录

CHAL: Causal-guided Hierarchical Anomaly-aware Learning for Moving Infrared Small Target Detection [CVPR 26]

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Fixing GRPO training collapse in long-horizon multi-tool agents. A lightweight PRM-Lite + LATA joint approach achieves +37% over vanilla GRPO on τ-bench airline (50-task, multi-turn).

[AAAI 26] SeViL: Semi-supervised Vision-Language Learning with Text Prompt Guiding for Moving Infrared Small Target Detection

A project implementing various agentic RL based on the Slime post-training framework

slime is an LLM post-training framework for RL Scaling.

🧠 Train a 64M-parameter LLM from scratch in just 2h!

https://hrl.boyuai.com/

Official implementation of "Diffuse to Detect: Bi-Level Sample Rebalancing with Pseudo-Label Diffusion for Point-Supervised Infrared Small-Target Detection", accepted to ICML 2026 Spotlight.

Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL

Official implementation of the paper "MIST: A Benchmark and Baseline for Multi-frame Infrared Small Target Detection in Complex Motion"

Code for paper V2V: Scaling Event-Based Vision through Efficient Video-to-Voxel Simulation.

[AAAI 2026 Oral] Official implementation for "I2E: Real-Time Image-to-Event Conversion for High-Performance Spiking Neural Networks"

Official implementation of "Towards Explainable Video Camouflaged Object Detection: SAM2 with Eventstream-Inspired Data" (AAAI 2026).

code for "Learning Global Dynamic Query for Large-Motion Infrared Small Target Detection"

上海大学硕博学位论文 LaTeX 模板 Thesis Template for Shanghai University

Fast Multi-dimensional Sparse Attention

MedSAM3: Delving into Segment Anything with Medical Concepts

🎓Automatically Update CV Papers Daily using Github Actions

Adapting Meta AI's Segment Anything to Downstream Tasks with Adapters and Prompts

EfficientSAM3 compresses SAM3 into lightweight, edge-friendly models via progressive knowledge distillation for fast promptable concept segmentation and tracking.

The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that sho…