Go to ICLR 2026 Conference homepageLearning Randomized Reductions
Keywords: Randomized Self-Reductions, Neuro-Symbolic Learning, Self-Correctness, Mathematical Function Learning
TL;DR: We automate the discovery of randomized self-reductions—mathematical properties that enable self-correcting programs—using both efficient regression and LLM-guided exploration.
Abstract: A self-corrector for a function $f$ takes a black-box oracle computing $f$ that is correct on most inputs and turns it into one that is correct on every input with high probability. Self-correctors exist for any function that is randomly self-reducible (RSR), where the value $f$ at a given point $x$ can be recovered by computing $f$ on random correlated points. While RSRs enable powerful self-correction capabilities and have applications in complexity theory and cryptography, their discovery has traditionally required manual derivation by experts. We present Bitween, a method and tool for automated learning of randomized self-reductions for mathematical functions. We make two key contributions: First, we demonstrate that our learning framework based on linear regression outperforms sophisticated methods including genetic algorithms, symbolic regression, and mixed-integer linear programming for discovering RSRs from correlated samples. Second, we introduce Agentic Bitween, a neuro-symbolic approach where large language models dynamically discover novel query functions for RSR property discovery, leveraging vanilla Bitween as a tool for inference and verification, moving beyond the fixed query functions ($x+r$, $x-r$, $x \cdot r$, $x$, $r$) previously used in the literature. On RSR-Bench, our benchmark suite of 80 scientific and machine learning functions, vanilla Bitween surpasses existing symbolic methods, while Agentic Bitween discovers new RSR properties using frontier models to uncover query functions.
Primary Area: neurosymbolic & hybrid AI systems (physics-informed, logic & formal reasoning, etc.)
Submission Number: 23040
Loading