仅用语义ID作为物品表示的推理重排序
原标题：Reasoning Reranking with Semantic IDs as the Sole Item Representation
AI 摘要
该论文提出 ReRankID 测试平台，研究仅用语义 ID（SID）作为物品表示进行推理重排序。实验发现现有模型在仅见 SID 时倾向于照抄输入顺序而非真正重排序，可读元数据也无法修复该失败。作者比较了监督微调、拒绝采样自训练、在线策略蒸馏和基于验证器的强化学习等后训练方法，发现它们能激发缺失的重排序行为；其中用分级排序奖励训练的 8B 模型在仅 SID 条件下表现最佳，匹配了提示长数倍的强文本阅读 LLM，并在长历史下保持增益。但配对理由审计显示，该模型虽更善于论证权衡，对物品的描述却不够准确，说明后训练提升了排序而非推理本身。
正文节选
Reasoning Reranking with Semantic IDs as the Sole Item Representation Abstract Reasoning rerankers write a rationale before ranking a candidate slate, and existing systems read item metadata to do so, even though generative recommenders increasingly represent items as compact Semantic IDs (SIDs). Ranking from SIDs alone has two appeals: SID prompts are far shorter than the matched item text, and although item codes are fixed from content, the backbone learns their token embeddings in part from u