ML Meanders
  • Posts

All Posts

  • The Thought That Counts: Exploring Chain of Thought Prompting

    February 1, 2026 · 9 min read

    We reproduce chain-of-thought experiments across base, instruction-tuned, and reasoning models. The results suggest that reported CoT gains on modern models are mostly artifacts of suppression, not necessarily reasoning improvements.

    • cot
  • On the Issues with Transpose Convolution and GAN Based Synthesis

    May 15, 2020 · 8 min read

    In this post, we discuss the issue of checkerboard artefacts introduced by transpose convolution layers as well as proposed solutions to this problem. Experiments on audio synthesis are also given, further motivating these methods.

    • deep-learning
    • gans
    • audio
    • speech-synthesis
  • Contrastive Predictive Coding for Representation Learning

    February 17, 2020 · 26 min read

    An exploration of Noise Contrastive Estimation (NCE) and how it enables Contrastive Predictive Coding (CPC) for learning useful representations from high-dimensional sequential data like speech.

    • deep-learning
    • representation-learning
    • self-supervised
    • audio

Following curiosity wherever it leads

RSS GitHub

© 2026 ML Meanders. Content licensed under CC BY 4.0.