Pinned Loading

1. DPO-for-Sarcasm-in-Small-Models DPO-for-Sarcasm-in-Small-Models Public

   Training efficient small language models for sarcasm detection through Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO).

   Jupyter Notebook

2. Insiders-Edge Insiders-Edge Public

   A recurrent dueling-DQN and PPO pipeline for partially observable stock trading with controllable signal-quality experiments.

   Jupyter Notebook 1

3. HalloweenHack HalloweenHack Public

   AI-powered Polymarket trading agent - built for the YC Halloween Hackathon, winner of the Hyperspell (YC F25) Prize.

   Python