| Reducing Hallucinations: Merlin-Arthur Training and Evaluation EXPLAINED | Sep 27, 2026 | 32:57 | 11.9K |
| Differential Privacy in 30 secondsShort | Nov 9, 2025 | 0:37 | 4.2K |
| Google's new VaultGemma model – Differential Privacy explained | Nov 2, 2025 | 8:37 | 6.4K |
| Flow Matching for text2image modelsShort | Oct 26, 2025 | 1:00 | 5.8K |
| Flow-Matching vs Diffusion Models explained side by side | Oct 19, 2025 | 16:08 | 58.2K |
| Energy-Based ModelsShort | Oct 12, 2025 | 0:45 | 4K |
| Energy-Based Transformers explained | How EBTs and EBMs work | Sep 21, 2025 | 14:35 | 21.1K |
| Inside ACL 2025 Vienna: Posters & Talks | Sep 14, 2025 | 9:12 | 3K |
| Min-p, the new kid on the LLM sampling blockShort | Aug 31, 2025 | 1:03 | 3.8K |
| Temperature Sampling in LLMsShort | Aug 24, 2025 | 0:50 | 3.4K |
| AI Word Choice Isn't Random – It’s Top-p SamplingShort | Aug 17, 2025 | 0:48 | 3K |
| LLMs don't Choose the Next Most Probable TokenShort | Aug 13, 2025 | 0:58 | 2.5K |
| Top-k sampling explainedShort | Aug 10, 2025 | 0:47 | 5.9K |
| Greedy? Min-p? Beam Search? How LLMs Actually Pick Words – Decoding Strategies Explained | Aug 3, 2025 | 11:53 | 8.2K |
| AlphaEvolve: Using LLMs to solve Scientific and Engineering Challenges | AlphaEvolve explained | Jun 19, 2025 | 8:58 | 7K |
| Token-Efficient Long Video Understanding for Multimodal LLMs | Paper explained | May 18, 2025 | 9:20 | 7.8K |
| 4-Bit Training for Billion-Parameter LLMs? Yes, Really. | Apr 18, 2025 | 15:40 | 14.9K |
| s1: Simple test-time scaling: Just “wait…” + 1,000 training examples? | PAPER EXPLAINED | Mar 23, 2025 | 6:09 | 5.6K |
| Only self-supervisionShort | Mar 2, 2025 | 1:03 | 1.7K |
| Training large language models to reason in a continuous latent space – COCONUT Paper explained | Jan 26, 2025 | 9:52 | 20.9K |
| Why do we need Tokenizers for LLMs?Short | Jan 22, 2025 | 1:11 | 2.3K |
| The best idea in NLP since 2017Short | Jan 21, 2025 | 0:58 | 2.5K |
| LLM Lecture: A Deep Dive into Transformers, Prompts, and Human Feedback | Jan 19, 2025 | 1:31:09 | 15.7K |
| 32 seconds of awkward silence with LetitiaShort | Jan 6, 2025 | 0:33 | 1.4K |
| REPA Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You ... | Dec 8, 2024 | 6:37 | 5.5K |
| Math Anxiety? and what it has to do with AI – 🔴at #HLF24 with Prof. Yael Tauman Kalai | Nov 3, 2024 | 7:52 | 4.1K |
| Graph Language Models EXPLAINED in 5 Minutes! [Author explanation 🔴 at ACL 2024] | Oct 6, 2024 | 6:38 | 7K |
| How OpenAI made o1 "think" – Here is what we think and already know about o1 reinforcement learning | Sep 13, 2024 | 9:24 | 11.8K |
| Transformer LLMs are Turing Complete after all !? | Sep 5, 2024 | 28:47 | 7.6K |
| Mission: Impossible language models – Paper Explained [ACL 2024 recording] | Sep 2, 2024 | 11:05 | 9.4K |
| Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution – Paper Explained | Aug 20, 2024 | 11:22 | 17.1K |
| My PhD Journey in AI / ML (while doing YouTube on the side) | Aug 16, 2024 | 37:18 | 12.7K |
| [Own work] On Measuring Faithfulness or Self-consistency of Natural Language Explanations | Jul 26, 2024 | 8:48 | 3.9K |
| Supercharging RAG with Generative Feedback Loops from Weaviate | Jun 17, 2024 | 11:08 | 6K |
| GaLore EXPLAINED: Memory-Efficient LLM Training by Gradient Low-Rank Projection | May 27, 2024 | 11:38 | 11K |
| Shapley Values Explained | Interpretability for AI models, even LLMs! | May 6, 2024 | 9:59 | 7K |
| Stealing Part of a Production LLM | API protects LLMs no more | Apr 8, 2024 | 18:49 | 18K |
| Genie explained 🧞 Generative Interactive Environments paper explained | Mar 4, 2024 | 9:22 | 4.8K |
| MAMBA and State Space Models explained | SSM explained | Feb 17, 2024 | 22:27 | 98.7K |
| Sparse LLMs at inference: 6x faster transformers! | DEJAVU paper explained | Feb 3, 2024 | 13:17 | 6.6K |
| Transformers explained | The architecture behind LLMs | Jan 21, 2024 | 19:48 | 41.8K |
| Direct Preference Optimization: Your Language Model is Secretly a Reward Model | DPO paper explained | Dec 22, 2023 | 8:55 | 44K |
| FunSearch from DeepMind explained | LLM hallucinations discover new math solutions!? | Dec 18, 2023 | 11:36 | 13.8K |
| DALL-E 3 is better at following Text Prompts! Here is why. — DALL-E 3 explained | Nov 5, 2023 | 8:03 | 4.5K |
| Adversarial Attacks and Defenses. The Dimpled Manifold Hypothesis. David Stutz from DeepMind #HLF23 | Oct 20, 2023 | 13:06 | 3.5K |
| What is LoRA? Low-Rank Adaptation for finetuning LLMs EXPLAINED | Sep 18, 2023 | 8:22 | 113.1K |
| Are ChatBots their own death? | Training on Generated Data Makes Models Forget – Paper explained | Aug 24, 2023 | 11:53 | 5.6K |
| The first law on AI regulation | The EU AI Act | Jul 30, 2023 | 14:37 | 10.9K |
| Say that 3 times in a row. 😅Short | Jul 24, 2023 | 0:05 | 2.1K |
| Author Interviews, Poster Highlights, Summary of the ACL 2023 Toronto NLP | Jul 24, 2023 | 50:36 | 4.6K |
