# From Noise to Signal: When Outliers Seed New Topics

arXiv: https://arxiv.org/abs/2603.18358 · HTML: https://mlsanity.com/paper/2603.18358/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 26 references that are in the map.
- First posted on arXiv 18 Mar 2026.

## Builds on (most cited first, 4 of 4)

- [BERTopic: Neural topic modeling with a class-based TF-IDF procedure](https://mlsanity.com/paper/2203.05794/) (2022)
- [The Dynamic Embedded Topic Model](https://mlsanity.com/paper/1907.05545/) (2019)
- [BERTrend: Neural Topic Modeling for Emerging Trends Detection](https://mlsanity.com/paper/2411.05930/) (2024)
- [Monitoring geometrical properties of word embeddings for detecting the emergence of new topics](https://mlsanity.com/paper/2111.03496/) (2021)

## Built on by (most cited first, 1 of 1)

- [Predicting Emerging Topics from Outliers: A Prospective Study of Weak Signals in Embedding Space](https://mlsanity.com/paper/2609.29183/) (2026)

## Abstract (arXiv)

Outliers in dynamic topic modeling are typically treated as noise, yet we show that some can serve as early signals of emerging topics. We introduce a temporal taxonomy of news-document trajectories that defines how documents relate to topic formation over time. It distinguishes anticipatory outliers, which precede the topics they later join, from documents that either reinforce existing topics or remain isolated. By capturing these trajectories, the taxonomy links weak-signal detection with temporal topic modeling and clarifies how individual articles anticipate, initiate, or drift within evolving clusters. We implement it in a cumulative clustering setting using document embeddings from eleven state-of-the-art language models and evaluate it retrospectively on HydroNewsFr, a French news corpus on the hydrogen economy. Inter-model agreement reveals a small, high-consensus subset of anticipatory outliers, increasing confidence in these labels. Qualitative case studies further illustrate these trajectories through concrete topic developments.