Engine finds changing views in political debates over time
TL;QR: Querying Temporal Discursive Patterns in Textual Corpora
Computation and LanguageComputers and SocietyDatabasesLogic in Computer Science
Summary
Large collections of texts often contain stories or opinions that change gradually over years, like politicians shifting their stance on an important topic. The authors created TL;QR, which is a special tool that can search for these changes by combining clever computer logic with the help of AI that understands language. Their system reads small pieces of text to judge their meaning, then carefully puts those pieces together to find long-term patterns without losing accuracy. They tested it on fake data and a big set of European Parliament speeches, successfully spotting shifts in political opinions about privacy laws.
What this means in practice
- •For policy analysts: Detect when political groups slowly change their stance on regulations by analyzing years of parliamentary speeches.
- •For legal compliance teams: Monitor long-term shifts in corporate or regulatory discourse to anticipate evolving compliance risks.
Authors
François Hublet, Vladimir Krsmanovic, Jesus Solano, Ahmed Bouhoula, Ana Oliveira da Costa
Abstract
Many questions about large textual corpora concern patterns that unfold over time, such as a political group whose stance on a topic drifts from support to criticism over several years of debates. We present TL;QR (Too Long; Query with logical temporal Reasoning), a declarative query language and open-source engine for finding such temporal discursive patterns. In TL;QR, a single query combines natural-language predicates, whose truth is judged by a Large Language Model (LLM), with regular expressions and the operators of a temporal logic. The LLM answers only local questions about individual text passages; a symbolic engine then aggregates these judgments over time and links every match to the recorded passages and predicate decisions behind it. With this division of labor, the temporal and relational reasoning stays exact and auditable even over long time horizons. In addition, when a query includes a regular-expression pre-filter, only a small fraction of the corpus ever reaches the LLM. On a suite of controlled synthetic benchmarks with temporally and semantically complex patterns, TL;QR is substantially more accurate than LLM-only baselines. In a case study on ten years of European Parliament debates (442,593 multilingual speeches), TL;QR identifies candidate shifts in the political groups' stances on the GDPR privacy regulation.