aikenkyu001/semantic_roundtrip_benchmark: Initial Release of the Semantic Round-trip Benchmark
Abstract
This release contains the full benchmark code, dataset structure, and materials for the paper: "A Semantic Round-trip Benchmark and Dataset for Probing the Generalization Gap in Language Models". Benchmark implementation Dataset generation pipeline 7,000+ trial log structure Paper PDF Prompts and evaluation scripts
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 3 January 2026