Papers by Jonathan Gratch
Social Influence Dialogue Systems: A Survey of Datasets and Models For Social Influence Tasks (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing research focuses on task-oriented or open-domain dialogue systems with influence skills. |
| Approach: | They propose to define and introduce a category of social influence dialogue systems that influence users’ cognitive and emotional responses. |
| Outcome: | The proposed system is task-oriented or goal-oriented, but it is not open-domain. |
CaSiNo: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing systems that negotiate with humans have broad applications in pedagogy and conversational AI. |
| Approach: | They propose to annotate persuasion strategies and perform correlation analysis to understand how dialogue behaviors are associated with the negotiation performance. |
| Outcome: | The proposed system improves negotiation performance for all strategies labeled as skewed . the proposed system is available on github.com/kushalchawla/ . |
KODIS: A Multicultural Dispute Resolution Dialogue Corpus (2025.naacl-long)
Copied to clipboard
| Challenge: | KODIS is a dyadic dispute resolution corpus containing thousands of dialogues from over 75 countries. |
| Approach: | They propose to use a dyadic dispute resolution corpus to examine how conflicts escalate through conversation rather than deal-making. |
| Outcome: | The proposed corpus contains thousands of dialogues from over 75 countries. |
Be Selfish, But Wisely: Investigating the Impact of Agent Personality in Mixed-Motive Human-Agent Interactions (2023.emnlp-main)
Copied to clipboard
| Challenge: | A natural way to design a negotiation dialogue system is via self-play RL: train an agent that learns to maximize its performance by interacting with a simulated user that has been designed to imitate human-human dialogue data. |
| Approach: | They propose to use RL to train an agent that learns to maximize its performance by interacting with a simulated user that has been designed to imitate human-human dialogue data. |
| Outcome: | The proposed system fails to learn the value of compromise in a negotiation, which can lead to no agreements, and ultimately hurt the model's overall performance. |
Mechanistic Interpretability of Emotion Inference in Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on large language models (LLMs) show promising capabilities in predicting human emotions from text. |
| Approach: | They investigate how autoregressive LLMs infer emotions by focusing on appraisal theory . they show that emotion representations are functionally localized to specific regions in the model . |
| Outcome: | The proposed model is functionally localized to specific regions in the model, and the results align with theoretical and intuitive expectations. |
Are LLMs Effective Negotiators? Systematic Evaluation of the Multifaceted Capabilities of LLMs in Negotiation Dialogues (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly being utilized as AI negotiation agents . however, prior research on LLMs lacks a systematic evaluation of their diverse capabilities in negotiation. |
| Approach: | They propose to analyze the multifaceted capabilities of Large Language Models (LLMs) across diverse dialogue scenarios throughout the stages of a typical negotiation interaction. |
| Outcome: | The proposed model outperforms GPT-4 in many negotiation tasks while identifying specific challenges, such as making subjective assessments and generating contextually appropriate, strategically advantageous responses. |
Psychological Steering in LLMs: An Evaluation of Effectiveness and Trustworthiness (2026.acl-long)
Copied to clipboard
Amin Banayeeanzade, Ala N. Tak, Fatemeh Bahrani, Anahita Bolourani, Leonardo Blas, Emilio Ferrara, Jonathan Gratch, Sai Praneeth Karimireddy
| Challenge: | Using a model with a high degree of emotion and personality control, large language models can be used to control socially interactive interactions. |
| Approach: | They propose a Psychologically-informed benchmark to evaluate LLM steering effectiveness and trustworthiness across emotion and personality domains. |
| Outcome: | The framework establishes the first holistic evaluation of emotion and personality steering, offering insights into its interpretability and reliability for socially interactive applications. |
ASTRA: A Negotiation Agent with Adaptive and Strategic Reasoning via Tool-integrated Action for Dynamic Offer Optimization (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing agents struggle due to bounded rationality in human data, low adaptability to counterpart behavior, and limited strategic reasoning. |
| Approach: | They propose a framework for turn-level offer optimization based on two core principles: opponent modeling and Tit-for-Tat reciprocity. |
| Outcome: | The proposed framework outperforms baselines across diverse partner agents and validates through human evaluation. |
The Niki and Julie Corpus: Collaborative Multimodal Dialogues between Humans, Robots, and Virtual Agents (L18-1)
Copied to clipboard
Ron Artstein, Jill Boberg, Alesia Gainer, Jonathan Gratch, Emmanuel Johnson, Anton Leuski, Gale Lucas, David Traum
| Challenge: | Niki and Julie corpus contains more than 600 dialogues between humans and robots . corpus includes audio and video recordings, results of ranking tasks, questionnaire responses . |
| Approach: | the corpus contains more than 600 dialogues between human participants and a robot . the dialogues are part of a collaborative item-ranking task designed to measure influence . |
| Outcome: | the corpus contains more than 600 dialogues between human participants and a robot or virtual agent . the dialogues contain conversational errors by the robot, which simulates typical of modern automated agents . |
Opponent Modeling in Negotiation Dialogues by Related Data Adaptation (2022.findings-naacl)
Copied to clipboard
| Challenge: | In a multi-issue negotiation, it involves inferring the relative importance that the opponent assigns to each issue under discussion, which is crucial for finding high-value deals. |
| Approach: | They propose a ranker for inferring the priority order of the opponent from partial dialogues without needing additional annotations for training. |
| Outcome: | The proposed model performs better than baselines while accessing fewer utterances from the opponent. |
Can Language Model Moderators Improve the Health of Online Discourse? (2024.naacl-long)
Copied to clipboard
Hyundong Cho, Shuai Liu, Taiwei Shi, Darpan Jain, Basem Rizk, Yuyang Huang, Zixun Lu, Nuan Wen, Jonathan Gratch, Emilio Ferrara, Jonathan May
| Challenge: | Existing efforts to automate conversational moderation have focused on banning harmful comments or deleting them, but such efforts can inadvertently push users towards echo chambers that exacerbate polarization. |
| Approach: | They propose a framework to assess models’ moderation capabilities independently of human intervention and propose 'conversational moderation' they propose to use language models as conversational moderators to provide specific feedback on toxic behavior but struggle to influence users to increase their levels of respect and cooperation. |
| Outcome: | The proposed framework assesses models’ moderation capabilities independently of human intervention and shows that appropriately prompted models provide specific and fair feedback on toxic behavior but struggle to influence users to increase their levels of respect and cooperation. |