{
  "id": 16116789,
  "title": "Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox",
  "url": "https://urgent.news/2026/10/22/turning-database-grading-into-rl-rewards-post-training-agents-with",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-22T22:24:34.000Z",
  "source": {
    "name": "Hugging Face",
    "slug": "hugging-face",
    "url": "https://huggingface.co/blog/microsoft/thinkingbox-training"
  },
  "original_language": "en",
  "account": null,
  "summary": "The article discusses the development of ThinkingBox, a tool that evaluates an AI agent's work by comparing the final database state with the expected outcome, and uses this as a reward function for training. The authors claim that this approach improved the performance of the Qwen3.8-27B model, increasing average success from 51.70% to 60.78% and the number of tasks completed correctly on all twenty attempts from 38 to 107. The improvement was particularly notable in a specific hotel-booking task, where the RL fine-tuned model preserved the booking and payment state in all 20 attempts, whereas the base model failed to do so in 15 attempts.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}