Q-TIE improves search results by understanding time in queries
Q-TIE: A Lightweight and Generalizable Re-ranking Framework for Temporal Information Retrieval
Computation and LanguageInformation Retrieval
Summary
Finding information relevant to a certain time is tricky because current methods either understand the query well or check time details well, but not both. The authors propose Q-TIE, which learns to find the time period a query refers to and uses that to better reorder search results. This helps to make sure the documents you see match both what you asked and the right time frame. Their tests show Q-TIE works better than previous methods and is easy to add to existing search systems.
What this means in practice
- •For search engine developers: Improve dynamic re-ranking of search results by incorporating learned temporal intent to match queries with time-relevant documents.
- •For digital assistant teams: Enhance response accuracy by filtering retrieved information to those documents aligned with the time constraints of user questions.
Authors
Soyeon Kim, Hyunjin Kim, JinYeong Bak, Steven Euijong Whang
Abstract
Temporal Information Retrieval (TIR) has been increasingly critical given the rise of Retrieval-Augmented Generation (RAG). Since temporally mismatched evidence can be highly misleading, TIR aims to retrieve documents that are both semantically and temporally relevant to a query. Two TIR paradigms have emerged - temporal retrievers and temporal re-rankers - differing in how temporal relevance is modeled. While these paradigms provide complementary strengths, our analysis reveals that each alone falls short of robust TIR: temporal retrievers provide flexible query understanding via learned representations, but often fail to explicitly account for temporal constraints; temporal re-rankers can enforce such constraints more explicitly, but often rely on predefined re-ranking rules. To address this, we propose Q-TIE, a re-ranking framework based on learned Temporal Intent Extraction (TIE). By introducing a TIE model that maps each query's temporal constraint into a unified interval representation (i.e., $\langle t_{start}, t_{end} \rangle$), Q-TIE generalizes beyond predefined rules via model-based learning while explicitly modeling temporal constraints as a separate signal - jointly achieving what each paradigm typically trades off. Experiments demonstrate that Q-TIE consistently outperforms existing TIR methods with stronger generalizability across temporal query types, and provides a lightweight yet effective add-on for temporally-aware RAG pipelines. Code: https://github.com/ssoy0701/Q-TIE.