Papers by Grace Hui Yang

4 papers
Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents (2026.findings-acl)

Copied to clipboard

Challenge: Existing systems for conversational AI are user-driven, but in many real-world situations, they do not extract information to achieve its own objectives.
Approach: They propose an inquisitive conversational agent that learns when and how to ask probing questions . they also propose a framework for a conversational ICA specifically tailored to the court .
Outcome: The proposed method outperforms single-agent RL baselines on a U.S. Supreme Court dataset.
YIELD: A Large-Scale Dataset and Evaluation Framework for Information Elicitation Agents (2026.acl-long)

Copied to clipboard

Challenge: Existing conversational agents (CAs) are designed to satisfy user needs through user-driven interactions. however, many real-world settings, such as academic interviewing, require agents that can elicit information from users.
Approach: They propose to support Information Elicitation Agents (IEAs) in which the agent’s goal is to elicit information from users to support the agent's institutional or task-oriented objectives.
Outcome: The proposed agent-based model improves the performance of a 26M-token dataset of 2,281 human-to-human dialogues on multiple foundation LLMs and human evaluation confirms the results.
More Diverse Dialogue Datasets via Diversity-Informed Data Collection (2020.acl-main)

Copied to clipboard

Challenge: Existing approaches to generate conversational dialogue produce uninteresting, predictable responses.
Approach: They propose a method to collect and determine more diverse data from conversational participants . they use dynamically computed corpus-level statistics to determine which conversational participant to collect data from .
Outcome: The proposed method produces significantly more diverse data than baseline methods and better results on emotion classification and dialogue generation tasks.
High-Quality Dialogue Diversification by Intermittent Short Extension Ensembles (2021.findings-acl)

Copied to clipboard

Challenge: Many task-oriented dialogue systems use deep reinforcement learning (DRL) to learn policies that respond to the user appropriately and complete the tasks successfully.
Approach: They propose a method to diversify dialogues using a set of user models by constraining the intensity to interact with diverse user models.
Outcome: The proposed method improves the performance of several state-of-the-art DRL dialogue agents trained in simulators.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations