A term-centric framework uses automatic term extraction to align heterogeneous document collections into a shared space and builds hierarchies by combining domain priors with clustering, outperforming document-level baselines on a 1M+ document English-German benchmark.
Topic Intrusion for Automatic Topic Model Evaluation
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
SBTA reformulates topic modeling to assign topics at the segment level rather than document level, yielding cleaner topics on a new SemEval-STM dataset created via LLM decomposition and human refinement.
citing papers explorer
-
Term-Centric Hierarchy Induction from Heterogeneous Corpora
A term-centric framework uses automatic term extraction to align heterogeneous document collections into a shared space and builds hierarchies by combining domain priors with clustering, outperforming document-level baselines on a 1M+ document English-German benchmark.
-
From Documents to Segments: A Contextual Reformulation for Topic Assignment
SBTA reformulates topic modeling to assign topics at the segment level rather than document level, yielding cleaner topics on a new SemEval-STM dataset created via LLM decomposition and human refinement.