t55
- Karma
- 0
- Created
- ()
- Submissions
- 0
- RL Speedrun (github.com)
- Target Policy Optimization (arxiv.org)
-
Hey HN — we’re a small team that uses Claude Code + Codex for basically everything in our company: coding, data analysis, marketing, ad campaigns, copywriting, design. There’s a truckload of tribal knowledge we’ve…
- Procedural Reasoning Datasets (github.com)
- In Defence of Gary Marcus (reubenadams.substack.com)
- Reasoning Gym – Procedural RL reasoning datasets (github.com)
- ChatGPT Agent [video] (youtube.com)
- Show HN: Rehearsal.so, Duolingo for Public Speaking (rehearsal.so)
- End-to-End Vision Tokenizer Tuning (arxiv.org)
- YC Interview Mock Practice (rehearsal.so)
- D1: Scaling Reasoning in Diffusion LLMs via Reinforcement Learning (dllm-reasoning.github.io)
- Are LLMs more than autocomplete? AI Debate (rehearsal.so)
- How to stay in flow while using Cursor or Windsurf (rehearsal.so)
- Generative Modelling in Latent Space (sander.ai)
- Show HN: Debate Uncle Bob – Is SQL Dead? (Voice RPG) (rehearsal.so)
- OpenAI O3 and O4-Mini (openai.com)
- Memory in ChatGPT (twitter.com)
- Superintelligence startup Reflection AI launches with $130M in funding (siliconangle.com)
- Novelty Left on the Table (ansatz.blog)