-
Fudan University
- Shanghai
-
19:58
(UTC +08:00) - https://www.fudan.edu.cn/
Highlights
- Pro
Stars
AutoPPT — AI-Powered Interactive Lecture Slides Generator | AI备课神器 | 一键生成一整周精美、可交互的课件!
A user-friendly & efficient knowledge distillation framework for LLMs, supporting off-policy, full vocab on-policy (OPD), cross-tokenizer, multimodal, and on-policy self-distillation.
A high-throughput and memory-efficient inference and serving engine for LLMs
Introducing XSafeClaw: The Open-Source Agent Safety Platform from Fudan University
Minimal reproduction of DeepSeek R1-Zero
LlamaIndex is the document processing platform for AI
🦄️ 🎃 👻 Clash Premium 规则集(RULE-SET),兼容 ClashX Pro、Clash for Windows 等基于 Clash Premium 内核的客户端。
Open-source red teaming framework for MLLMs with 42+ attack methods
Awesome-Backdoor-on-LMMs is a collection of state-of-the-art, novel, exciting backdoor methods on LMMs (VLPs, TDMs, VLMs, and Agents).
😎 up-to-date & curated list of awesome Attacks on Large-Vision-Language-Models papers, methods & resources.
BackdoorVLM: A Benchmark for Backdoor Attacks on Vision-Language Models
EvoSynth is a SOTA automated LLM red-teaming framework that evolves executable, code-level jailbreak attack methods using multi-agent synthesis, evolutionary search, and self-correction.
✨ [ICLR'26] WithAnyone is capable of generating high-quality, controllable, and ID consistent images
Code for visual (PGD) and textual (Optimized HotFlip) attacker against OpenFlamingo. You will need to modify the openflamingo source code to make better use of this attacker.
Code for ICCV2025 paper——IDEATOR: Jailbreaking and Benchmarking Large Vision-Language Models Using Themselves
verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework
An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)
An educational resource to help anyone learn deep reinforcement learning.
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety
[NeurIPS 2025] BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models
Code for ACM MM2024 paper: White-box Multimodal Jailbreaks Against Large Vision-Language Models
deeplearning.ai(吴恩达老师的深度学习课程笔记及资源)