# Beyond Accuracy: Interpreting Topic Representation in Suicide Ideation Detection Models

arXiv: https://arxiv.org/abs/2606.07714 · HTML: https://mlsanity.com/paper/2606.07714/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 76 references that are in the map.
- First posted on arXiv 5 Jun 2026.

## Builds on (most cited first, 18 of 18)

- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [Snorkel: Rapid Training Data Creation with Weak Supervision](https://mlsanity.com/paper/1711.10160/) (2017)
- [Concept Whitening for Interpretable Image Recognition](https://mlsanity.com/paper/2002.01650/) (2020)
- [Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet](https://mlsanity.com/paper/2605.29358/) (2026)
- [Concept-based Explainable Artificial Intelligence: A Survey](https://mlsanity.com/paper/2312.12936/) (2023)
- [Overlooked factors in concept-based explanations: Dataset choice, concept learnability, and human capability](https://mlsanity.com/paper/2207.09615/) (2022)
- [Efficient Dictionary Learning with Switch Sparse Autoencoders](https://mlsanity.com/paper/2410.08201/) (2024)
- [Anchored Correlation Explanation: Topic Modeling with Minimal Domain Knowledge](https://mlsanity.com/paper/1611.10277/) (2016)
- [Challenges in Applying Explainability Methods to Improve the Fairness of NLP Models](https://mlsanity.com/paper/2206.03945/) (2022)
- [Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness](https://mlsanity.com/paper/2410.21331/) (2024)
- [ChatGPT for Suicide Risk Assessment on Social Media: Quantitative Evaluation of Model Performance, Potentials and Limitations](https://mlsanity.com/paper/2306.09390/) (2023)
- [Towards Procedural Fairness: Uncovering Biases in How a Toxic Language Classifier Uses Sentiment Information](https://mlsanity.com/paper/2210.10689/) (2022)
- [Concept-Based Explanations to Test for False Causal Relationships Learned by Abusive Language Classifiers](https://mlsanity.com/paper/2307.01900/) (2023)

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

- [Validation-Gated Causal Interventions for Interpreting High-Stakes Large Language Model Behavior: A Case Study in Suicidality Detection](https://mlsanity.com/paper/2606.21078/) (2026)

## Abstract (arXiv)

Suicide ideation detection models are typically evaluated using aggregate performance metrics, yet little is known about how they internally represent psychologically meaningful risk factors. In high-stakes mental health applications, understanding these internal representations is essential for safety, transparency, and responsible deployment. In this work, we move beyond accuracy and analyze how suicide detection models trained on original and topic-augmented datasets encode psychological risk factors in their internal representation space. Using visualization and geometric analysis, we examine the coherence and separability of topic-related features. Our results show that topic-aware augmentation increases the clarity and distinctness of underrepresented psychosocial risk factors such as immigration, family issues, and financial crisis. These findings suggest that augmentation not only improves model performance but also leads to more structured and interpretable internal representations.