AutoIndexer：通过编辑链训练因果模型实现灵活顺序解码
原标题：AutoIndexer: Flexible-Order Decoding by Training Causal Models on Chains of Edits
AI 摘要
Autodesk AI Lab 提出 AutoIndexer，一种通过「编辑链」训练因果模型实现灵活顺序解码的方法。模型学习预测插入、替换、删除等编辑操作的标记与光标位置，从而在生成后对已有输出进行回溯修改，且不牺牲标准 LLM 基准性能。该方法在生成顺序上保持自回归（保留 KV 缓存与 EOS 提前停止优势），但在显示顺序上可非自回归，代码已开源。
正文节选
BeNTo: Beyond Next-Token Prediction – Diffusion & Flow Models for Next-Generation Decoding AutoIndexer: Flexible-Order Decoding by Training Causal Models on Chains of Edits Abstract Autoregressive decoding is less suited to non-causally ordered data, e.g., cross-references or future dependencies, since their outputs cannot be revised post hoc, unlike iterative human reasoning. Diffusion LLMs partially mitigate this limitation but still cannot flexibly insert or remove arbitrary numbers of tokens