🚀 2026届大模型算法岗实习面经 | 包含 DeepSeek/Qwen 技术报告解析、手撕 PPO/RoPE/Transformer、RLHF 核心与八股文 | 持续更新中...

🌟100+ 原创 LLM / RL 原理图📚，《大模型算法》作者巨献！💥（100+ LLM/RL Algorithm Maps ）

DeepSeek 系列工作解读、扩展和复现。

MedicalGPT: Training Your Own Medical GPT Model with ChatGPT Training Pipeline. 训练医疗大模型，实现了包括增量预训练(PT)、有监督微调(SFT)、RLHF、DPO、ORPO、GRPO。

customized-qwen2.5vl-sft/grpo

MiniGPT-Pancreas: Multimodal Large language Model for Pancreas Cancer Classification and Detection

deep tensor canonical correlation analysis

Wrapper for a PyTorch classifier which allows it to output prediction sets. The sets are theoretically guaranteed to contain the true class with high probability (via conformal prediction).

Ongoing research training transformer models at scale

[ICCV'25 Highlight] Derm1M: A Million‑Scale Vision‑Language Dataset Aligned with Clinical Ontology Knowledge for Dermatology

ChatDev 2.0: Dev All through LLM-powered Multi-Agent Collaboration

Official repository for the ACL 2025 Findings paper "Worse than Random? An Embarrassingly Simple Probing Evaluation of Large Multimodal Models in Medical VQA"

PMC-VQA is a large-scale medical visual question-answering dataset, which contains 227k VQA pairs of 149k images that cover various modalities or diseases.

HealthCards: Exploring Text-to-Image Generation as Visual Aids for Healthcare Knowledge Democratizing and Education

MC-CoT implementation code

[MICCAI'25] Reliable few-shot transfer of medical VLMs using a novel transductive split conformal adaptation solver.

Official repository of Expert-Controlled Classifier-Free Guidance for Reliable Medical Visual Question Answering.

[IEEE Transactions on Medical Imaging/TMI 2023] This repo is the official implementation of "LViT: Language meets Vision Transformer in Medical Image Segmentation"

复现大模型相关算法及一些学习记录

Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

Reliable Conflictive Multi-view Learning

[TNNLS'25] [MICCAI'24] A Parameter and Memory Efficient Transfer Learning Method

医学影像数据集列表 『An Index for Medical Imaging Datasets』

Official code repository of paper HMIL: Hierarchical Multi-Instance Learning for Fine-Grained WSI Classification

Project 3 for my EECS 281 class. This project focused on working with hash tables and managing and creating larger data structures through composition by simulating a bank.

https://eecs281staff.github.io/p3-281bank/

A PyTorch toolbox for domain generalization, domain adaptation and semi-supervised learning.

[MICCAI 2024 Early Accept, Oral] Aligning Medical Images with General Knowledge from Large Language Models