A HyDE retrieval pipeline with full-answer context and adaptive similarity thresholding improves LLM answers to Stack Overflow questions over zero-shot prompting for three of four open-source models.
Knowledgeable or Educated Guess? Revisiting Language Models as Knowledge Bases
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Previous literatures show that pre-trained masked language models (MLMs) such as BERT can achieve competitive factual knowledge extraction performance on some datasets, indicating that MLMs can potentially be a reliable knowledge source. In this paper, we conduct a rigorous study to explore the underlying predicting mechanisms of MLMs over different extraction paradigms. By investigating the behaviors of MLMs, we find that previous decent performance mainly owes to the biased prompts which overfit dataset artifacts. Furthermore, incorporating illustrative cases and external contexts improve knowledge prediction mainly due to entity type guidance and golden answer leakage. Our findings shed light on the underlying predicting mechanisms of MLMs, and strongly question the previous conclusion that current MLMs can potentially serve as reliable factual knowledge bases.
citation-role summary
background 1
citation-polarity summary
fields
cs.SE 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Never Come Up Empty: Adaptive HyDE Retrieval for Improving LLM Developer Support
A HyDE retrieval pipeline with full-answer context and adaptive similarity thresholding improves LLM answers to Stack Overflow questions over zero-shot prompting for three of four open-source models.