PinnedBuilding BlocksinTowards AIComparing Dense Attention vs Sparse Sliding Window AttentionPart 2: LLMs May Not Need Dense Self AttentionDec 24, 2023Dec 24, 2023
PinnedBuilding BlocksLLMs May Not Need Dense Self AttentionSink Tokens and the Sparsity of Attention Scores in Transformer ModelsNov 11, 2023Nov 11, 2023
PinnedBuilding BlocksinTowards AIPaper Review: Multimodal Chain of Thought ReasoningLanguage Models improve with Visual FeaturesFeb 7, 20231Feb 7, 20231
PinnedBuilding BlocksVisualizing Attacking Build-Up Play Using Dynamic Passing NetworksLeveraging Probabilistic Sampling, K-Nearest Neighbors and Longest Common SubsequenceOct 20, 20221Oct 20, 20221
Building BlocksPaper Review: What is UL2 in Flan-UL2?A Mixture of Denoisers and Mode SwitchingMar 13, 2023Mar 13, 2023
Building BlocksVisual ChatGPT: Paper and Code ReviewPowering ChatGPT with Visual Foundation ModelsMar 12, 20232Mar 12, 20232
Building BlocksinTowards AIIs AI Becoming the Gatekeeper and Mouthpiece of Knowledge?AI vs Human Expertise: The Growing Dependence on Machine-Learned KnowledgeMar 8, 20231Mar 8, 20231
Building BlocksPart 2: Enjoying the Subtler Aspects of FootballPressing, Baiting, AmbipedalJan 30, 2023Jan 30, 2023
Building BlocksPaper Review: Constituional AI, Training LLM’s using PrinciplesGoverning the behavior of Generative AI through PrinciplesJan 26, 2023Jan 26, 2023