Here, we collect awesome AI/NLP/LLM learning resource (Conference Tutorials, Open Courses, etc.). Feel free to star and fork.
Any comments, suggestions? Let us know! We love PRs :)
Explanation
π means the official website of the resource.
Publicly available course Slides, Notes (Chinese Translation Maybe), Assignments resources and Readings information.
- CS224N: Natural Language Processing with Deep Learning - Natural language processing (NLP) or computational linguistics is one of the most important technologies of the information age. Applications of NLP are everywhere because people communicate almost everything in language: web search, advertising, emails, customer service, language translation, virtual agents, medical reports, politics, etc. In the 2010s, deep learning (or neural network) approaches obtained very high performance across many different NLP tasks, using single end-to-end neural models that did not require traditional, task-specific feature engineering. In the 2020s amazing further progress was made through the scaling of Large Language Models, such as ChatGPT. In this course, students will gain a thorough introduction to both the basics of Deep Learning for NLP and the latest cutting-edge research on Large Language Models (LLMs). Through lectures, assignments and a final project, we will learn the necessary skills to design, implement, and understand our own neural network models, using the Pytorch framework. π
- Word Vectors
- Word Vectors and Language Models
- Python Review Session
- Backpropagation and Neural Network Basics
- Dependency Parsing
- PyTorch Tutorial Session
- Basic Sequence Models to RNNs
- Advanced Variants of RNNs, Attention
- Transformers
- Pretraining
- Post-training (RLHF, SFT, DPO)
- Hugging Face Transformers Tutorial Session
- Efficient Adaptation (Prompting + PEFT)
- Benchmarking and Evaluation
- Question Answering and Knowledge
- Assignment
- 11-667: Large Language Models - Methods and Applications - A graduate-level course that aims to provide a holistic view of the current state of LLMs. The first half of this course starts with the basics of language models, including network architectures, training, inference, and evaluation. Then it discusses the interpretation (or attempts of), alignments, and emergent capabilities of LLMs, followed by its popular applications in language tasks and new utilizations beyond texts. In the second half, this course first presents the techniques of scaling up language model pretraining and recent approaches in making the pretraining of LLMs and their deployment more efficient. It then discusses various concerns surrounding the deployment of LLMs and wraps up with the challenges and frontiers of recent developments. The course is designed to give us an overview of the techniques behind LLMs and a thorough grounding on the fundamentals and cutting-edge developments of LLMs, to prepare us for further research or applied endeavors in this domain. π
- Building blocks of modern LLMs
- Transformer architecture and pre-training learning
- Pre-training data curation and tokenization
- Architecture Advancement on Transformers
- Automatic evaluation of LLMs
- In-context learning, task-oriented finetuning, and parameter-efficient tuning methods
- Should we expect a half of a translation model to reason?
- Evaluation and Learning in Human-Agent Interaction
- Surprising Behaviours of In-Context Learning and Model Alignment
- Retrieval-augmented generation
- Scaling Up LLMs - Scaling laws
- Scaling up LLMs - Optimization and Parallel Training
- Interpretability methods
- Bias and ethical issues
- Attacking LLM systems + benchmark leakage
- LLMs for Recommendation Systems
- Reinforcement Learning from Human Feedback (RLHF)
- Efficient Inference Methods
- Chatbots and LLM Agents
- Long-context models
- Efficient Pretraining with Sparse Models
- Training with Synthetic Data
- Assignment
Publicly available AI top conference (CCF A/B) tutorials part Slides resources and Readings List.
- AAAI 2025
- Half Day
- Bridging Inverse Reinforcement Learning and Large Language Model Alignment - Toward Safe and Human-Centric AI Systems
- Building trustworthy ML - The role of label quality and availability
- Fairness in AI:ML via Social Choice
- Foundation Models meet Embodied Agents
- Multi-modal Foundation Model for Scientific Discovery - With Applications in Chemistry, Material, and Biology
- Pre-trained Language Model with Limited Resources
- Concept-based Interpretable Deep Learning
- Evaluating Large Language Models - Challenges and Methods
- Foundation Models for Time Series Analysis - A Tutorial
- Neurosymbolic AI for EGI - Explainable, Grounded, and Instructable Generations
- (Really) Using Counterfactuals to Explain AI Systems - Fundamentals, Methods, & User Studies for XAI
- Advancing Brain-Computer Interfaces with Generative AI for Text, Vision, and Beyond
- AI for Science in the Era of Large Language Models
- Causal Representation Learning
- Graph Neural Networks - Architectures, Fundamental Properties and Applications
- Machine Learning for Protein Design
- The Lifecycle of Knowledge in Large Language Models - Memorization, Editing, and Beyond
- Thinking with Functors β Category Theory for A(G)I
- User-Driven Capability Assessment of Taskable AI Systems
- AI Data Transparency - The Past, the Present, and Beyond
- Data-driven Decision-making in Public Health and its Real-world Applications
- Decision Intelligence for Two-sided Marketplaces
- Inferential Machine Learning - Towards Human-collaborative Vision and Language Models
- Machine Learning for Solvers
- Model Reuse - Unlocking the Power of Pre-Trained Model Resources
- Symbolic Regression - Towards Interpretability and Automated Scientific Discovery
- Tutorial - Multimodal Artificial Intelligence in Healthcare
- Half Day Lab
- Quarter Day
- Advancing Offline Reinforcement Learning - Essential Theories and Techniques for Algorithm Developers
- Unified Semi-Supervised Learning with Foundation Models
- Reinforcement Learning with Temporal Logic objectives and constraints
- Deep Representation Learning for Tabular Data
- LLMs and Copyright Risks - Benchmarks and Mitigation Approaches
- Physics-Inspired Geometric Pretraining for Molecule Representation
- From Tensor Factorizations to Circuits (and Back)
- KV Cache Compression for Efficient Long Context LLM Inference - Challenges, Trade-Offs, and Opportunities
- Supervised Algorithmic Fairness in Distribution Shifts
- Artificial Intelligence Safety - From Reinforcement Learning to Foundation Models
- Hallucinations in Large Multimodal Models
- Graph Machine Learning under Distribution Shifts - Adaptation, Generalization and Extension to LLM
- Curriculum Learning in the Era of Large Language Models
- Hypergraph Neural Networks - An In-Depth and Step-by-Step Guide
- The Quest for A Science of Language Models
- When Deep Learning Meets Polyhedral Theory - A Tutorial
- Quarter Day Labs
- SOFAI Lab - A Hands-On Guide to Building Neurosymbolic Systems with Metacognitive Control
- Continual Learning on Graphs - Challenges, Solutions, and Opportunities
- Developing explainable multimodal AI models with hands-on lab on the life-cycle of rare event prediction in manufacturing
- Financial Inclusion through AI-Powered Document Understanding
- Half Day
- ACL 2025
- Inverse Reinforcement Learning Meets Large Language Model Alignment
- Eyetracking and NLP
- Uncertainty Quantification for Large Language Models
- Human-AI Collaboration - How AIs Augment Human Teammates
- Navigating Ethical Challenges in NLP - Hands-on strategies for students and researchers
- NLP for Counterspeech against Hate and Misinformation
- Synthetic Data in the Era of Large Language Models
- Guardrails and Security for LLMs - Safe, Secure, and Controllable Steering of LLM Applications
- COLING 2025
- Speculative Decoding for Efficient LLM Inference
- From Theory to Practice - A Hands-on Tutorial in Explainable NLP
- Hands-On Tutorial - Labeling with LLM and Human-in-the-Loop
- LLMs in Education - Novel Perspectives, Challenges, and Opportunities
- EduRAG - Crafting Clever Educational Chatbots and Advanced QA Systems with RetrievalAugmented Generation
- Connecting Ideas in Lower-Resource Scenarios - NLP for National Varieties, Creoles, and Other Low-Resource Scenarios
- Safety Issues for Generative AI
- Hallucinative Foundation Models - Characterization, Quantification, Avoidance, and Mitigation
- Bridging Linguistic Theory and AI - Usage-Based Learning in Humans and Machines
- EMNLP 2025
- Efficient Inference for Large Language Models β Algorithm, Model, and System π
- Advancing Language Models through Instruction Tuning - Recent Progress and Challenges π
- Spoken Conversational Agents with Large Language Models π
- NLP+Code - Code Intelligence in Language Models π
- Data and Model Centric Approaches for Expansion of Large Language Models to New languages π
- Neuro-Symbolic Natural Language Processing π
- Continual Learning of Large Language Models π
- ICML 2025
- Game-theoretic Statistics and Sequential Anytime-Valid Inference
- Calibration and Bias in Algorithms, Data, and Models - a tutorial on metrics and plots for measuring calibration, bias, fairness, reliability, and robustness
- Modern Methods in Associative Memory
- Alignment Methods for Large Language Models
- Generative AI Meets Reinforcement Learning
- Tutorial on Mechanistic Interpretability for Language Models
- Harnessing Low Dimensionality in Diffusion Models - From Theory to Practice
- DP-fy your DATA - How to (and why) synthesize Differentially Private Synthetic Data
- Jailbreaking LLMs and Agentic Systems - Attacks, Defenses, and Evaluations
- Flowing Through Continuous-Time Generative Models - A Clear and Systematic Tour
- Training Neural Networks at Any Scale
- The Underlying Logic of Language Models
- IJCAI 2025
- Scaling LLM Training - Efficient Pre-training & Fine-tuning on AI Accelerators
- LLM-based Role-Playing from the Perspective of Hallucinations
- Evaluating LLM-based Agents - Foundations, Best Practices and Open Challenges
- Empowering LLMs with Logical Reasoning - Challenges, Solutions, and Opportunities
- Large Language Models for Recommendation
- Beyond Text - Advanced Retrieval Augmented Generation for Complex and Multimodal Data
- Neuroevolution of Intelligent Agents
- AI Meets Algebra - Foundations and Frontiers
- Multimodal Large Language Model for Visually Rich Document Understanding
- Principles of Self-supervised Learning in the Foundation Model Era
- Gradient-Based Multi-Objective Deep Learning
- Advances in Time-Series Anomaly Detection
- GUI Agents with Foundation Models - Data Resource, Framework and Application
- Federated Compositional and Bilevel Optimization
- Multi-Modal Generative AI in Dynamic and Open Environment
- A Tutorial on Bandit Learning in Matching Markets
- Deep Learning for Graph Anomaly Detection
- Towards Low-Distortion Graph Representation Learning
- Beyond Graph Distribution Shifts - LLMs, Adaptation, and Generalizati
- Supervised Algorithmic Fairness in Distribution Shifts
- Fairness in Large Language Models - A Tutorial
- Human-Centric and Multimodal Evaluation for Explainable AI - Moving Beyond Benchmarks
- Computational Pathology Foundation Models - Datasets, Adaptation Strategies, and Evaluations
- NAACL 2025
- Creative Planning with Language Models - Practice, Evaluation and Applications
- DAMAGeR - Deploying Automatic and Manual Approaches to GenAI Red-teaming
- Foundation Models Meet Embodied Agents
- Knowledge Distillation for Language Models
- Adaptation of Large Language Models
- Learning Language through Grounding
- LLMs and Copyright Risks - Benchmarks and Mitigation Approaches
- Social Intelligence in the Age of LLMs
- NeurIPS 2025
- Mexico City
- Efficient Transformers - State of the art in pruning, sparse attention, and transformer funneling
- How to Build Agents to Generate Kernels for Faster LLMs (and Other Models!)
- Geospatial Foundation Models - Overview, Application and Benchmarking
- From Tuning to Guarantees - Statistically Valid Hyperparameter Selection
- Science of Trustworthy Generative Foundation Models π
- Positional Encoding - Past, Present, and Future π
- San Diego
- Autoregressive Models Beyond Language
- Data Privacy, Memorization, & Legal Implications in Generative AI - A Practical Guide π
- Energy and Power as First-Class ML Design Metrics π
- Foundations of Imitation Learning - From Language Modeling to Continuous Control π
- Foundations of Tensor:Low-Rank Computations for AI
- Human-AI Alignment - Foundations, Methods, Practice, and Challenges π
- Model Merging - Theory, Practice and Applications
- New Frontiers of Hyperparameter Optimization - Recent advances and open challenges in theory and practice
- Scale Test-Time Compute on Modern Hardware
- Planning in the Era of Language Models π
- Theoretical Insights on Training Instability in Deep Learning π
- The Science of Benchmarking - Whatβs Measured, Whatβs Missed, and Whatβs Next π
- Explain AI Models - Methods and Opportunities in Explainable AI, Data-Centric AI, and Mechanistic Interpretability π
- Recent Developments in Geometric Machine Learning - Foundations, Models, and More
- Mexico City