Papers by Mengxiang Li

2 papers
Dual Prompt Tuning based Contrastive Learning for Hierarchical Text Classification (2024.findings-acl)

Copied to clipboard

Challenge: Existing methods focus on hierarchy-aware text feature by exploiting explicit parent-child relationships, resulting in label confusion within each layer.
Approach: They propose a dual-prompt tuning method which emphasizes discrimination among peer labels by performing contrastive learning on each hierarchical layer.
Outcome: The proposed method outperforms existing methods on benchmark datasets and is available on github.
Beyond Static Personas: Situational Personality Steering for Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Existing personalization methods rely on static personality modeling to achieve optimal performance.
Approach: They propose a training-free framework for advanced situational personality steering that incorporates situation-dependent behavior patterns within LLM personalities through analysis of persona neurons.
Outcome: The proposed framework surpasses baselines on PersonalityBench and SPBench, demonstrating generalization and robustness to complex, unseen situations and different models architecture.

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