| --- |
| title: README |
| emoji: 🐢 |
| colorFrom: yellow |
| colorTo: gray |
| sdk: static |
| pinned: false |
| license: mit |
| papers: |
| - 2601.21839 |
| --- |
| |
| # Human-Centric Machine Learning |
| ### Max Planck Institute for Software Systems |
|
|
| ## 📚 Featured Research |
|
|
| **Test-Time Compute Games** |
| *Strategic behavior in LLM-as-a-service markets.* |
| * 📄 [Read the Paper](https://arxiv.org/abs/2601.21839) |
| * 💻 [View Code (GitHub)](https://github.com/Human-Centric-Machine-Learning/strategic-ttc) |
| * 💾 [**Download Dataset**](https://huggingface.co/datasets/Human-Centric-Machine-Learning/strategic-ttc-data) — *Contains full inference traces: sampled answers, token counts, correctness labels, and Best-of-N scores.* |
|
|
| **Tokenization Multiplicity Leads to Arbitrary Price Variation in LLM-as-a-service** |
| *Solving tokenization multiplicity with a simple and efficient sampling method.* |
| * 📄 [Read the Paper](https://www.arxiv.org/abs/2506.06446) |
| * 💻 [View Code (GitHub)](https://github.com/Human-Centric-Machine-Learning/Tokenization-Multiplicity) |
| * 💾 [**Download Dataset**](https://huggingface.co/datasets/Human-Centric-Machine-Learning/tokenization-multiplicity-data) |
