Multimodal Large Language Models (MLLM)

Rules & FAQ

Class recordings (in Russian)

YouTube

VKVideo

Class Materials

#	Date	Title	Materials
1	Feb 11	Word Embeddings and Classification & Language Modelling	slides
		Embeddings & CNN/LSTM LMs with PyTorch	notebook
2	Feb 18	Seq2seq, Attention, and Transformers	slides
		Transformer from Scratch	notebook
3	Feb 25	Pretraining, SFT, RLHF & PEFT, LoRA	slides
		Parameter-efficient fine-tuning	notebook
4	Mar 4	Reasoning, RLVF & RAG	slides
		Tokenization	notebook
6	Mar 18	Introduction to MLLMs and Image Modality	slides
		Classification of VLMs: Deep Fusion vs Early Fusion	notebook
7	Mar 25	VLLM and Data Generation	slides
		Visual Autoregressive Transformer	notebook
8	Apr 1	Video Understanding	slides
		Video Modality and Any-to-any Models	notebook
9	Apr 8	Action Modality (Robotics)	slides
		Intro to Vision Language Action Models	notebook
10	Apr 15	Intelligent Document Processing (IDP) и UI Agents	slides
		Agentic Workflow	notebook
11	Apr 22	3D Data Modality	slides
		VLM Grounding	notebook
12	Apr 29	Efficient Inference: FlashAttention, KV cache, Distillation, Quantization	slides
		KV cache, Quantization	notebook