Dynamic Topic Modeling for Cross-Corpus Temporal Analysis
2026-08-24 • Computation and Language
Computation and Language
AI summaryⓘ
The authors improve how dynamic embedded topic models (D-ETM) compare topics over time across different text collections. Instead of training topics separately for each collection and matching them later, they first create a shared topic space from all collections combined. Then, each collection can fine-tune the topics slightly without changing the overall shared topics, making comparisons more reliable. Their method works better than previous ways, keeping topics aligned over nearly 100 years of data from three different text sources.
Dynamic Embedded Topic Model (D-ETM)Topic modelingTemporal semantic evolutionCross-corpus comparisonTopic alignmentResidual adaptationCorpus of Historical American EnglishHungarian matchingTrajectory retrieval
Authors
Ruoxuan Li, Bruce Kogut
Abstract
Dynamic Embedded Topic Models (D-ETM) provide an interpretable framework for modeling temporal semantic evolution, but cross-corpus comparison remains difficult because topics are often learned independently and aligned only after training, a process that does not guarantee stable topic correspondence across corpora and time. To address this problem, we propose a D-ETM framework that first learns a common dynamic topic space over a merged multi-corpus collection, which we call the shared backbone, then introduces corpus-specific residual adaptation around the frozen backbone without creating separate latent topic spaces. This design preserves a shared topic index for cross-corpus comparison while allowing each corpus to specialize lexically. We evaluate the framework on three temporally structured corpora spanning 97 years: the Corpus of Historical American English, Harvard Business Review, and International Labour Review. Residual adaptation improves corpus-specific fit relative to the shared backbone while preserving the same-index cross-corpus topic trajectories, achieving substantially stronger alignment than full fine-tuning from the same backbone, with $97.5 \pm 0.7\%$ versus $17.9 \pm 1.1\%$ trajectory Retrieval@1, as well as stronger alignment than independent training with post-hoc Hungarian matching. These results suggest that incorporating topic alignment into the model can support more stable over-time cross-corpus comparisons while retaining corpus-specific lexical variation.