Understanding AI alignment faking AI alignment occurs when an AI system performs exactly the function it was designed for—e.g., reading and summarizing documen...
Determining a winner among a set of items using active pairwise comparisons under a limited budget is a challenging problem in preference-based learning. The go...
Decision Trees The unreasonable power of nested decision rules. By Jared Wilberhttps://twitter.com/jdwlbr & Lucía Santamaríahttps://twitter.com/lusantala Let's...
Article URL: https://alexlitzenberger.com/blog/post.html?post=/building_a_minimal_transformer_for_10_digit_addition Comments URL: https://news.ycombinator.com/i...
We develop a discrete gauge-theoretic framework for superposition in large language models (LLMs) that replaces the single-global-dictionary premise with a shea...
This paper presents a controlled empirical study of biologically motivated local learning for handwritten digit recognition. We evaluate an STDP-inspired compet...
Scaling video generation from seconds to minutes faces a critical bottleneck: while short-video data is abundant and high-fidelity, coherent long-form data is s...
The fast-growing demands in using Large Language Models (LLMs) to tackle complex multi-step data science tasks create an emergent need for accurate benchmarking...
Multi-turn interactions with large language models typically retain the assistant's own past responses in the conversation history. In this work, we revisit thi...
GPU kernel optimization is fundamental to modern deep learning but remains a highly specialized task requiring deep hardware expertise. Despite strong performan...
Modern optimizers like Adam and Muon are central to training large language models, but their reliance on first- and second-order momenta introduces significant...
Transformers have been established as the de-facto backbones for most recent advances in sequence modeling, mainly due to their growing memory capacity that sca...
