Erdős problems and equational laws
Open statements settled by a Lean proof or a finite counterexample (placeholder)
Open statements settled by a Lean proof or a finite counterexample (placeholder)
Sequences that fold into hard target structures and into nothing else (placeholder)
Codon sequences that fold stably and express well, under manufacturing constraints (placeholder)
Buildable coils whose field matches a target plasma boundary (placeholder)
Fewer T gates on standard circuits, and stabilizer codes with larger distance (placeholder)
Step functions and finite sets that tighten autocorrelation and sum–difference constants (placeholder)
Train GPT-2 small to a fixed validation loss in as little time as possible (placeholder)
Predict the fitness effects of mutations across 217 deep mutational scans (placeholder)
Which hypothetical inorganic crystals are thermodynamically stable (placeholder)
Faster GPU kernels for PyTorch operators, exactly as correct as the reference (placeholder)
Lesion classification on multi-centre breast MRI, over seven cross-validation folds (placeholder)
Neural surrogates that roll out PDE solutions with less error (placeholder)
Faster scientific-computing functions that still match SciPy’s output (placeholder)
More Putnam competition problems proved in Lean 4 within a fixed search budget (placeholder)