REVIEW 10 cited by
Top2Vec: Distributed Representations of Topics
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Topic modeling is used for discovering latent semantic structure, usually referred to as topics, in a large collection of documents. The most widely used methods are Latent Dirichlet Allocation and Probabilistic Latent Semantic Analysis. Despite their popularity they have several weaknesses. In order to achieve optimal results they often require the number of topics to be known, custom stop-word lists, stemming, and lemmatization. Additionally these methods rely on bag-of-words representation of documents which ignore the ordering and semantics of words. Distributed representations of documents and words have gained popularity due to their ability to capture semantics of words and documents. We present $\texttt{top2vec}$, which leverages joint document and word semantic embedding to find $\textit{topic vectors}$. This model does not require stop-word lists, stemming or lemmatization, and it automatically finds the number of topics. The resulting topic vectors are jointly embedded with the document and word vectors with distance between them representing semantic similarity. Our experiments demonstrate that $\texttt{top2vec}$ finds topics which are significantly more informative and representative of the corpus trained on than probabilistic generative models.
Forward citations
Cited by 10 Pith papers
-
Unsupervised Multimodal Graph-based Model for Geo-social Analysis
A joint text-and-location graph model with contrastive, coherence, and alignment losses produces topic clusters that are semantically coherent and spatially compact on four disaster tweet datasets.
-
SEADialogues: A Multilingual Culturally Grounded Multi-turn Dialogue Dataset on Southeast Asian Languages
SEADialogues is a culturally grounded multi-turn dialogue dataset covering eight Southeast Asian languages.
-
Conceptual Topic Aggregation
FAT-CAT builds a directory-level concept lattice from topic-model outputs, adding image captions and cross-folder aggregation to earlier FCA-based topic analysis.
-
Constrained Non-negative Matrix Factorization for Guided Topic Modeling of Minority Topics
A constrained NMF with a single seed word list and prevalence constraints improves detection of low-prevalence topics, at least on a small synthetic benchmark.
-
Improving Topic Modeling of Social Media Short Texts with Rephrasing: A Case Study of COVID-19 Related Tweets
LLM rephrasing of tweets before topic modeling raises Wikipedia-measured coherence (LDA 0.31→0.50) but the abstract's claim of broad improvements is contradicted by the paper's own table for LDA and by the metric choice.
-
Hybrid Topic-Semantic Labeling and Graph Embeddings for Unsupervised Legal Document Clustering
Concatenating Top2Vec and Node2Vec embeddings, where the Node2Vec graph encodes Top2Vec's own topic labels, yields compact clusters, but the gain is largely circular.
-
Granite Embedding R2 Models
Granite Embedding R2 is an Apache-2.0 family of ModernBERT-based retrieval and reranking models that posts high average scores on several benchmarks but falls short of top code-retrieval and reranking baselines.
-
GeoMoE: Divide-and-Conquer Motion Field Modeling with Mixture-of-Experts for Two-View Geometry
The paper as submitted does not contain the GeoMoE method or experiments, so the stated two-view geometry result is unverifiable.
-
Disaster Informatics after the COVID-19 Pandemic: Bibliometric and Topic Analysis based on Large-scale Academic Literature
A bibliometric study of 5,494 disaster informatics papers (2020-2022) finds the U.S., India, and UK lead in output, and identifies 12 dominant topics including public health themes.
-
FinBERT2: A Specialized Bidirectional Encoder for Bridging the Gap in Finance-Specific Deployment of Large Language Models
A 32B-token Chinese financial corpus and FinBERT2 model outperform prior FinBERTs, general BERTs, and several large LLMs on five classification and retrieval benchmarks.
Discussion (0). Sign in to comment.