In plain words: The model teaches itself by making easier versions of hard problems and solving them, with no human feedback or curated data. On undergraduate calculus integration, a small model went from solving almost nothing to 82% of problems.
Abstract · LADDER: Self-Improving LLMs Through Recursive Problem Decomposition
We introduce LADDER (Learning through Autonomous Difficulty-Driven Example Recursion), a framework which enables Large Language Models to autonomously improve their problem-solving capabilities through self-guided learning by recursively generating and solving progressively simpler variants of complex problems. Unlike prior approaches that require curated datasets or human feedback, LADDER leverages a model's own capabilities to generate easier question variants. We demonstrate LADDER's effectiveness in the subject of mathematical integration, improving Llama 3.2 3B's accuracy from 1% to 82% on undergraduate-level problems and enabling Qwen2.5 7B Deepseek-R1 Distilled to achieve 73% on the MIT Integration Bee qualifying examination. We also introduce TTRL (Test-Time Reinforcement Learning), where we perform reinforcement learning on variants of test problems at inference time. TTRL enables Qwen2.5 7B Deepseek-R1 Distilled to achieve a state-of-the-art score of 90% on the MIT Integration Bee qualifying examination, surpassing OpenAI o1's performance. These results show how self-directed strategic learning can achieve significant capability improvements without relying on architectural scaling or human supervision.
Toby Simonds, Akira Yoshiyama
arXiv:2503.00735 · cs.LG, cs.AI · submitted Mar 2, 2025 · updated Mar 5, 2025
abstract · pdf · html
I have seen at least 3 interesting/mildly promising breakthroughs on ML just these past two days! I mean, a Google research team just discovered that you can combine NNs with CLAs using digital logic gates as a medium, so you could potentially reduce many kinds of non-linear problems to a simple, efficient digital circuit! And it was on the HN front page, TODAY![1]
I keep seeing more mind-bending stuff related to neural nets and logic/intelligence in general, my mind has been running wild with speculation about the future and just how close we could (or could not) be to truly understanding how intelligence works from first principles.
[1] https://news.ycombinator.com/item?id=43286161