{"slug": "mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting", "title": "MIT Researchers Used GPT-6 Astra To Edit Qwen's Open Weights Directly, Boosting Math Scores", "summary": "Researchers at MIT's Computer Science & Artificial Intelligence Laboratory (CSAIL) reported on November 2 that OpenAI's GPT-6 Astra raised the MATH-500 score of the open-source Qwen3.5-0.8B model from 7.9% to 48.1% by directly editing approximately 0.016% of its weights, an approach the team calls \"weight-space programming.\" The work, described by Rikiya Takehi, Jayden Teoh, Alex Zhang and Omar Khattab in a research blog post, altered the network without conventional training procedures, with the gain coming primarily from correcting repetitive reasoning. Voyage AI staff researcher Chandan Singh said critics of the post are \"actually *underselling* the approach.", "body_md": "An anonymous reader quotes a report from VentureBeat: Researchers from the Massachusetts Institute of Technology (MIT) Computer Science & Artificial Intelligence Laboratory (CSAIL) reported today that they used OpenAI's GPT-6 Astra language model to raise a small, open source Qwen model's math score roughly sixfold by directly editing its weights -- the numerical parameters that govern its behavior. Rikiya Takehi, Jayden Teoh, Alex Zhang and Omar Khattab described their work in a research blog post published November 2. They call the approach \"weight-space programming\" and tested Astra's ability to interpret and alter networks without using conventional training procedures to generate the edits.\n \nIn the strongest language-model example, Astra's intervention raised Qwen3.5-0.8B's MATH-500 score from 7.9% to 48.1% while touching approximately 0.016% of its weights. The improvement came primarily from correcting repetitive reasoning—a distinction that shaped both the technical findings and the debate over their implications. Their experiments explored a broader possibility: AI systems that inspect, repair and construct other neural networks by working directly on their internal machinery. \"This is cool as heck!\" wrote Yuchen Jin, a technical staffer at Databricks and former co-founder and CTO of Hyperbolic Labs who was unaffiliated with the research, on X. \"Imagine using a large LLM to directly edit a smaller model's weights and dramatically improve its capabilities. It's like gene editing for neural networks. Even cooler than distillation!\"\n \nVoyage AI staff researcher Chandan Singh, formerly of Microsoft AI, also reacted to the findings: \"For anyone questioning whether this post is overhyped, I think they are actually *underselling* the approach.\"\n[Read more of this story](https://science.slashdot.org/story/26/11/03/0537252/mit-researchers-used-gpt-6-astra-to-edit-qwens-open-weights-directly-boosting-math-scores?utm_source=rss1.0moreanon&utm_medium=feed) at Slashdot.", "url": "https://wpnews.pro/news/mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting", "canonical_source": "https://science.slashdot.org/story/26/11/03/0537252/mit-researchers-used-gpt-6-astra-to-edit-qwens-open-weights-directly-boosting-math-scores?utm_source=rss1.0mainlinkanon&utm_medium=feed", "published_at": "2026-11-03 16:00:00+00:00", "updated_at": "2026-11-03 16:19:27.392571+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["Massachusetts Institute of Technology", "MIT Computer Science & Artificial Intelligence Laboratory", "OpenAI", "GPT-6 Astra", "Qwen3.5-0.8B", "Rikiya Takehi", "Omar Khattab", "Chandan Singh"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting", "markdown": "https://wpnews.pro/news/mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting.md", "text": "https://wpnews.pro/news/mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting.txt", "jsonld": "https://wpnews.pro/news/mit-researchers-used-gpt-6-astra-to-edit-qwen-s-open-weights-directly-boosting.jsonld"}}