# LinguIUTics at PsyDefDetect: Iterative Imbalance-Aware Fine-tuning of Qwen3-8B for Psychological Defense Mechanism Classification

Source: [LinguIUTics at PsyDefDetect: Iterative Imbalance-Aware Fine-tuning of Qwen3-8B for Psychological Defense Mechanism Classification](<https://www.sota2.com/research/paper/6a1e63c7dbd7150506f853d1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-02T05:00:51.557Z

Authors: Shefayat E Shams Adib, Ahmed Alfey Sani, Md Hasibur Rahman Alif, Ajwad Abrar

Published: 2026-05-30T09:47:29.000Z

## Abstract

Detecting psychological defense mechanisms in conversational text remains a challenging clinical NLP problem. For the PsyDefDetect 2026 shared task (nine-class utterance classification evaluated via macro F1), our team LinguIUTics achieves a macro F1-score of 0.3917 on the official positive-class leaderboard, ranking 4th out of 21 registered teams and improving over the Ministral-8B task baseline (31.48 macro F1) by 7.7 absolute points (24.4 percent relative). BERT-family encoders and zero-shot LLMs proved ineffective on rare classes due to severe class imbalance, leading us to QLoRA fine-tuning of Qwen3-8B. We leverage three key strategies: grouped stratified cross-validation (preventing leakage), minority-class round-robin lexical augmentation, and a post-processing pipeline with logit bias tuning and ensemble blending. Together, these components close much of the validation-to-leaderboard gap and substantially improve minority-class recall, driving the critical "Unclear" class (Level 8) from near-zero performance to an F1 score of 0.797.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Psychological Defense Classification on PsyDefConv 2026 (official leaderboard)](<https://www.sota2.com/research/sota/psychological-defense-classification-on-psydefconv-2026-official-leaderboard.md>): Psychological Defense Classification; PsyDefConv 2026 (official leaderboard); Accuracy: 64.19.

- [Psychological Defense Detection on PsyDefDetect (test)](<https://www.sota2.com/research/sota/psychological-defense-detection-on-psydefdetect-test.md>): Psychological Defense Detection; PsyDefDetect (test); Accuracy: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.00647>)

[PDF](<https://arxiv.org/pdf/2606.00647.pdf>)