對話式推薦系統零資料引導之實證研究

摘要

對話式推薦系統（CRS）通常需要特定領域的對話資料，但這類資料成本高昂、數量稀缺，且在新領域中往往無法取得。我們針對零資料 CRS 自舉訓練（zero-data CRS bootstrapping）進行了系統性的實證研究：從非對話訊號——商品評論、元資料與使用者-商品互動——生成合成對話監督訊號，且完全不使用任何領域內對話語料庫。我們比較了兩種基於資訊理論的選擇策略——Jensen-Shannon 多樣性與 Fisher 資訊——並跨領域訊號、模型架構、資料集與微調範式進行評估。結果顯示，以領域為根基的合成資料 consistently outperform zero-shot prompting and naive synthetic baselines。具體而言，主動選擇在資料效率上優於隨機抽樣；元資料與協同過濾訊號各自都能提升選擇品質；而在低資源情境下，合成資料不僅能勝過稀缺的真實對話，還能進一步與其產生互補。這些發現確立了非對話領域訊號作為建構無對話訓練資料之 CRS 的可行途徑。程式碼可於 https://anonymous.4open.science/r/zero_data_crs/ 取得。

English

Conversational Recommender Systems (CRS) typically require domain-specific dialogue data, which is costly, scarce, and often unavailable in new domains. We conduct a systematic empirical study of zero-data CRS bootstrapping: generating synthetic conversational supervision from non-conversational signals---item reviews, metadata, and user-item interactions---without any in-domain dialogue corpus. We compare two information-theoretic selection strategies, Jensen-Shannon diversity and Fisher information, across domain signals, model architectures, datasets, and fine-tuning paradigms. Our results show that domain-grounded synthetic data consistently outperforms zero-shot prompting and naive synthetic baselines; active selection improves data efficiency over random sampling; metadata and collaborative filtering signals each improve selection quality; and, in low-resource settings, synthetic data can outperform scarce real dialogues while further complementing them. These findings establish non-conversational domain signals as a viable path toward building CRS without conversational training data. The code is available at https://anonymous.4open.science/r/zero_data_crs/ .