困惑度陷阱：基於預訓練語言模型的檢索器過度高估低困惑度文檔

摘要

先前的研究發現，基於預訓練語言模型（PLM）的檢索模型對大型語言模型（LLM）生成的內容表現出偏好，即使這些文檔的語義質量與人類撰寫的相當，仍會給予更高的相關性評分。這一現象被稱為來源偏差，威脅著信息獲取生態系統的可持續發展。然而，來源偏差的根本原因尚未被探討。在本文中，我們通過因果圖解釋了信息檢索的過程，並發現基於PLM的檢索器學習了困惑度特徵來進行相關性估計，從而通過將低困惑度的文檔排名更高來引發來源偏差。理論分析進一步揭示，這一現象源於語言建模任務和檢索任務中損失函數梯度之間的正相關性。基於此分析，我們提出了一種因果啟發的推理時間去偏方法，稱為因果診斷與校正（CDC）。CDC首先診斷困惑度的偏差效應，然後將偏差效應從整體估計的相關性評分中分離出來。跨三個領域的實驗結果展示了CDC卓越的去偏效果，強調了我們提出的解釋框架的有效性。源代碼可在https://github.com/WhyDwelledOnAi/Perplexity-Trap獲取。

English

Previous studies have found that PLM-based retrieval models exhibit a preference for LLM-generated content, assigning higher relevance scores to these documents even when their semantic quality is comparable to human-written ones. This phenomenon, known as source bias, threatens the sustainable development of the information access ecosystem. However, the underlying causes of source bias remain unexplored. In this paper, we explain the process of information retrieval with a causal graph and discover that PLM-based retrievers learn perplexity features for relevance estimation, causing source bias by ranking the documents with low perplexity higher. Theoretical analysis further reveals that the phenomenon stems from the positive correlation between the gradients of the loss functions in language modeling task and retrieval task. Based on the analysis, a causal-inspired inference-time debiasing method is proposed, called Causal Diagnosis and Correction (CDC). CDC first diagnoses the bias effect of the perplexity and then separates the bias effect from the overall estimated relevance score. Experimental results across three domains demonstrate the superior debiasing effectiveness of CDC, emphasizing the validity of our proposed explanatory framework. Source codes are available at https://github.com/WhyDwelledOnAi/Perplexity-Trap.

困惑度陷阱：基於預訓練語言模型的檢索器過度高估低困惑度文檔

Perplexity Trap: PLM-Based Retrievers Overrate Low Perplexity Documents

摘要

Summary

Support

Support