基于意图的语义ID生成用于 grounded 对话新闻推荐
Intent-Driven Semantic ID Generation for Grounded Conversational News Recommendation
AI总结：
本文提出意图驱动的语义ID生成方法，通过多任务对齐和GPT-4链式推理蒸馏，实现意图到层级SID前缀的映射，提升新闻推荐的 grounded 性和冷启动用户推荐效果。
摘要：
对话新闻推荐需要将每个建议 grounding 在快速演变的文章语料库中，同时处理缺乏显式可检索关键词的隐含用户意图。为此，我们从生产对话中识别出6种意图类型：五种是隐含的，对标准RAG流程构成根本性挑战，形成关键的检索优先瓶颈。为解决这些问题，我们引入意图驱动的语义ID（SID）生成，采用生成后再匹配的范式。通过两阶段训练，包括多任务SID对齐和GPT-4链式推理蒸馏，LLM将多样化的意图映射到层级SID前缀，然后通过模糊匹配当前新闻池以确保完全 grounded 的推荐。Profile-Aware Dual-Signal Reasoning（PADR）进一步使冷启动用户仅通过资料库即可获得有效推荐。在主流中文新闻平台上，我们的7B模型在152K开放生成SID空间中实现0%幻觉和12.4%L1匹配（4倍随机基线）。它在L1指标上与GPT-4+混合RAG相当，在更细粒度的指标（L2 2倍，类别+1.2pp）上以约100倍更低的成本超越。冷启动用户，现有基线得分为0%，达到18.0%L1（6倍随机），在所有用户群体中最高。
英文摘要：
Conversational news recommendation requires grounding each suggestion in a rapidly evolving article corpus while addressing implicit user intents that lack explicit retrievable keywords. To characterize this scenario, we identify 6 intent types from production dialogues: five are implicit and pose fundamental challenges to standard RAG pipelines, forming a critical retrieve-first bottleneck. To address these issues, we introduce intent-driven Semantic ID (SID) generation under a Generate-then-Match paradigm. With two-stage training that consists of multi-task SID alignment and GPT-4 Chain-of-Thought distillation, an LLM maps diverse intents to hierarchical SID prefixes, which are then fuzzy-matched to the current news pool to guarantee fully grounded recommendations. Profile-Aware Dual-Signal Reasoning (PADR) further enables cold-start users to obtain valid recommendations using only profiles. On a mainstream Chinese news platform, our 7B model achieves 0% hallucination and 12.4% L1 match in the 152K open-generation SID space (4x random baseline). It matches GPT-4+Hybrid RAG on L1 while surpassing it on finer-grained metrics (L2 2x, Category +1.2pp) at ~100x lower cost. Cold-start users, where existing baselines score 0%, achieve 18.0% L1 (6x random), the highest among all user groups.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
