Hou, B., Jiang, G., Quan, G., Li, W., Feng, W., Liu, G., & Zhang, Y. (2026). Rethinking Cross-Tokenizer On-Policy Distillation: From Alignment Coverage to Supervision Reliability. https://omanscience.com/en/articles/rethinking-cross-tokenizer-on-policy-distillation-from-alignment-coverage-to-supervision-reliability