# Evaluating Contextual Embeddings and their Extraction Layers for Depression Assessment

arXiv: https://arxiv.org/abs/2112.13795 · HTML: https://mlsanity.com/paper/2112.13795/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 33 references that are in the map.
- First posted on arXiv 27 Dec 2021.

## Builds on (most cited first, 11 of 11)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://mlsanity.com/paper/2002.08909/) (2020)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [NRC-Canada: Building the State-of-the-Art in Sentiment Analysis of Tweets](https://mlsanity.com/paper/1308.6242/) (2013)
- [Empirical Evaluation of Pre-trained Transformers for Human-Level NLP: The Role of Sample Size and Dimensionality](https://mlsanity.com/paper/2105.03484/) (2021)
- [MeLT: Message-Level Transformer with Masked Document Representations as Pre-Training for Stance Detection](https://mlsanity.com/paper/2109.08113/) (2021)
- [Residualized Factor Adaptation for Community Social Media Prediction Tasks](https://mlsanity.com/paper/1808.09479/) (2018)

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

- [Systematic Evaluation of GPT-3 for Zero-Shot Personality Estimation](https://mlsanity.com/paper/2306.01183/) (2023)

## Abstract (arXiv)

Recent works have demonstrated ability to assess aspects of mental health from personal discourse. At the same time, pre-trained contextual word embedding models have grown to dominate much of NLP but little is known empirically on how to best apply them for mental health assessment. Using degree of depression as a case study, we do an empirical analysis on which off-the-shelf language model, individual layers, and combinations of layers seem most promising when applied to human-level NLP tasks. Notably, we find RoBERTa most effective and, despite the standard in past work suggesting the second-to-last or concatenation of the last 4 layers, we find layer 19 (sixth-to last) is at least as good as layer 23 when using 1 layer. Further, when using multiple layers, distributing them across the second half (i.e. Layers 12+), rather than last 4, of the 24 layers yielded the most accurate results.