mlsanityData to 14 Oct 2026

The Pursuit of Empathy: Evaluating Small Language Models for PTSD Dialogue Support

arXiv 2505.15065 ↗First posted 21 May 2025cs.CL cs.AI cs.CYcited by 6

Not on the map. Why?

Abstract (arXiv)

This paper investigates the capacity of small language models (0.5B-5B parameters) to generate empathetic responses for individuals with PTSD. We introduce Trauma-Informed Dialogue for Empathy (TIDE), a novel dataset comprising 10,000 two-turn conversations across 500 diverse, clinically-grounded PTSD personas (https://huggingface.co/datasets/yenopoya/TIDE). Using frontier model outputs as ground truth, we evaluate eight small LLMs in zero-shot settings and after fine-tuning. Fine-tuning enhances empathetic capabilities, improving cosine similarity and perceived empathy, although gains vary across emotional scenarios and smaller models exhibit a "knowledge transfer ceiling." As expected, Claude Sonnet 3.5 consistently outperforms all models, but surprisingly, the smaller models often approach human-rated empathy levels. Demographic analyses showed that older adults favored responses that validated distress before offering support (p = .004), while graduate-educated users preferred emotionally layered replies in specific scenarios. Gender-based differences were minimal (p > 0.15), suggesting the feasibility of broadly empathetic model designs. This work offers insights into building resource-efficient, emotionally intelligent systems for mental health support.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 22 of its 67 references that are in the map. First posted on arXiv 21 May 2025.

Builds on

22 of 67 references.

References · 22

  1. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  2. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,077
  3. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 515
  4. 2019On Measuring Social Biases in Sentence Encoderscited by 398
  5. 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 264
  6. 2020A Computational Approach to Understanding Empathy Expressed in Text-Based Mental Health Supportcited by 182
  7. 2021Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approachcited by 75
  8. 2021CoMAE: A Multi-factor Hierarchical Framework for Empathetic Response Generationcited by 42
  9. 2023Harnessing the Power of Large Language Models for Empathetic Response Generation: Empirical Investigations and Improvementscited by 42
  10. 2023Empowering Psychotherapy with Large Language Models: Cognitive Distortion Detection through Diagnosis of Thought Promptingcited by 31
  11. 2023Chain of Empathy: Enhancing Empathetic Response of Large Language Models Based on Psychotherapy Modelscited by 24
  12. 2023Harnessing Large Language Models' Empathetic Response Generation Capabilities for Online Mental Health Counselling Supportcited by 17
10 more
  1. 2024Enhancing Empathetic Response Generation by Augmenting LLMs with Small-scale Empathetic Modelscited by 13
  2. 2025Thousand Voices of Trauma: A Large-Scale Synthetic Dataset for Modeling Prolonged Exposure Therapy Conversationscited by 11
  3. 2024Multi-dimensional Evaluation of Empathetic Dialog Responsescited by 10
  4. 2022Privacy Sensitive Speech Analysis Using Federated Learning to Assess Depressioncited by 8
  5. 2025How Real Are Synthetic Therapy Conversations? Evaluating Fidelity in Prolonged Exposure Dialoguescited by 8
  6. 2024A Novel Nuanced Conversation Evaluation Framework for Large Language Models in Mental Healthcited by 7
  7. 2022Alexa as an Active Listener: How Backchanneling Can Elicit Self-Disclosure and Promote User Experiencecited by 6
  8. 2024Toward Large Language Models as a Therapeutic Tool: Comparing Prompting Techniques to Improve GPT-Delivered Problem-Solving Therapycited by 5
  9. 2024Enhancing Mental Health Support through Human-AI Collaboration: Toward Secure and Empathetic AI-enabled chatbotscited by 5
  10. 2024Towards a Multidimensional Evaluation Framework for Empathetic Conversational Systemscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 5.

6 papers.

  1. 2026PERM: Psychology-grounded Empathetic Reward Modeling for Large Language Modelscited by 5
  2. 2025Fine-Tuning Large Audio-Language Models with LoRA for Precise Temporal Localization of Prolonged Exposure Therapy Elementscited by 3
  3. 2026Responsible Evaluation of AI for Mental Healthcited by 2
  4. 2026AI Safety Training Can be Clinically Harmfulcited by 1
  5. 2026From Empathy to Personalized Empathy: Adapting Empathetic Strategies to Individual Userscited by 0
  6. 2026Towards High-Level Semantic Intelligencecited by 0