SciDesignBench: Benchmarking and Improving Language Models for Scientific Inverse Design

ArXi:2603.12724v1 Announce Type: new Many of the most important problems in science and engineering are inverse problems: given a desired outcome, find a design that achieves it. Evaluating whether a candidate meets the spec is often routine; a binding energy can be computed, a reactor yield simulated, a pharmacokinetic profile predicted. But searching a combinatorial design space for inputs that satisfy those targets is fundamentally harder. We