# Steering Conversational Large Language Models for Long Emotional Support Conversations

arXiv: https://arxiv.org/abs/2402.10453 · HTML: https://mlsanity.com/paper/2402.10453/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 23 references that are in the map.
- First posted on arXiv 16 Feb 2024.

## Builds on (most cited first, 20 of 20)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Steering Language Models With Activation Engineering](https://mlsanity.com/paper/2308.10248/) (2023)
- [Towards Emotional Support Dialog Systems](https://mlsanity.com/paper/2106.01144/) (2021)
- [A Computational Approach to Understanding Empathy Expressed in Text-Based Mental Health Support](https://mlsanity.com/paper/2009.08441/) (2020)
- [SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF](https://mlsanity.com/paper/2310.05344/) (2023)
- [Do Long-Range Language Models Actually Use Long-Range Context?](https://mlsanity.com/paper/2109.09115/) (2021)
- [MISC: A MIxed Strategy-Aware Model Integrating COMET for Emotional Support Conversation](https://mlsanity.com/paper/2203.13560/) (2022)
- [Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approach](https://mlsanity.com/paper/2101.07714/) (2021)
- [RankGen: Improving Text Generation with Large Ranking Models](https://mlsanity.com/paper/2205.09726/) (2022)
- [Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning](https://mlsanity.com/paper/2210.04242/) (2022)
- [Control Globally, Understand Locally: A Global-to-Local Hierarchical Graph Network for Emotional Support Conversation](https://mlsanity.com/paper/2204.12749/) (2022)
- [From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning](https://mlsanity.com/paper/2310.00492/) (2023)
- [Steering Large Language Models for Machine Translation with Finetuning and In-Context Learning](https://mlsanity.com/paper/2310.13448/) (2023)
- [TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition](https://mlsanity.com/paper/2305.03296/) (2023)
- [FADO: Feedback-Aware Double COntrolling Network for Emotional Support Conversation](https://mlsanity.com/paper/2211.00250/) (2022)

## Built on by (most cited first, 8 of 8)

- [ESC-Eval: Evaluating Emotion Support Conversations in Large Language Models](https://mlsanity.com/paper/2406.14952/) (2024)
- [HEART: A Unified Benchmark for Assessing Humans and LLMs in Emotional Support Dialogue](https://mlsanity.com/paper/2601.19922/) (2026)
- [ESC-Judge: A Framework for Comparing Emotional Support Conversational Agents](https://mlsanity.com/paper/2505.12531/) (2025)
- [You Never Know a Person, You Only Know Their Defenses: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations](https://mlsanity.com/paper/2512.15601/) (2025)
- [Empathy Through Multimodality in Conversational Interfaces](https://mlsanity.com/paper/2405.04777/) (2024)
- [Affective Flow Language Model for Emotional Support Conversation](https://mlsanity.com/paper/2602.08826/) (2026)
- [Do We Talk to Robots Like Therapists, and Do They Respond Accordingly? Language Alignment in AI Emotional Support](https://mlsanity.com/paper/2506.16473/) (2025)
- [Chain-of-Conceptual-Thought Elicits Daily Conversation in Large Language Models](https://mlsanity.com/paper/2510.18434/) (2025)

## Abstract (arXiv)

In this study, we address the challenge of enabling large language models (LLMs) to consistently adhere to emotional support strategies in extended conversations. We focus on the steerability of the Llama-2 and Llama-3 suite of models, examining their ability to maintain these strategies throughout interactions. To assess this, we introduce the Strategy Relevant Attention (SRA) metric, which quantifies the model's adherence to the prompted strategy through attention maps. To facilitate our study, we create a strategy-conditioned synthetic conversational dataset derived from the ESConv dataset. We also propose various baselines informed by our proposed SRA metric to address the challenge and propose a fine-tuned model that significantly enhances the steerability of the base model in following the strategy throughout the conversation. The code and data are publicly available on our GitHub.