• [Article] ChatGPT for mental health.

  • [Article] AI singer Xania Monet became the first known AI artist to debut on a Billboard radio chart.

  • [Article] Mercor raises another $350M at $10B valuation.

  • [Paper] Recursive reasoning in latent space on extremely small models to solve Sudoku & other analytical tasks.

  • [Paper] The Markovian Thinker - context managemenet for thinking tokens as a way to condense & solidify policy prior for RL.

  • [Paper] Aligning visual & language latent spaces & other cross-modal properties as a cost-effective way to scale up vision-language primitives.

  • [Paper] The Free Transformer. Architectural variation of the standard transformer. Insert an encoder in the middle of the decoder to make it explicitly condition on latent variables.

  • [Blog] Thinking Machines’ writeup on on-policy distillation. Combine the on-policy relevance of RL with the dense reward signal of distillation.

Have a great week!

Updated:

Comments