7 papers
Apodex Discovery: Reality Benchmarks and Environments for Evaluating and Building Discoverative Artificial Intelligence
Brian Wang, Bin Feng, Xiaoman Pan +26
Apollo did not reach the Moon merely because its engineers could solve difficult equations. It succeeded by turning a distant ambition into a mission architecture of explicit objec…
QED: An Open-Source Multi-Agent System for Generating Mathematical Proofs on Open Problems
Chenyang An, Qihao Ye, Minghao Pan +1
We present QED, an open-source multi-agent system that turns human-provided research questions into complete mathematical proofs without further human guidance. Its pipeline is des…
Return Probability for the Switch--Walk--Switch Lamplighter Walk on a Regular Tree
Chenyang An, Minghao Pan
We derive the sharp return-probability asymptotic for the switch--walk--switch lamplighter walk with lamp group over the infinite -regular tree: \[ p_{2n}(e,e) = Ï…
Lower Bounds for Advection-Diffusion Equations: An Exploration with AI-Generated Proofs
Chenyang An, Xiaoqian Xu
We establish explicit lower bounds for advection-diffusion equations in three settings: a polynomial bound for inviscid shears with , a uni…
Next-Token Prediction Task Assumes Optimal Data Ordering for LLM Training in Proof Generation
Chenyang An, Shima Imani, Feng Yao +8
In the field of large language model (LLM)-based proof generation, despite extensive training on large datasets such as ArXiv, LLMs still exhibit only modest performance on proving…
The Price of Format: Diversity Collapse in LLMs
Longfei Yun, Chenyang An, Zilong Wang +2
Instruction-tuned large language models (LLMs) employ structured templates, such as role markers and special tokens, to enforce format consistency during inference. However, we ide…