Papers by Shuzhen Li

4 papers
Incongruity-aware Tension Field Network for Multi-modal Sarcasm Detection (2025.acl-long)

Copied to clipboard

Challenge: Multi-modal sarcasm detection (MSD) identifies sarcasm and accurately understands users’ real attitudes from text-image pairs.
Approach: They propose to use incongruity-aware tension field network to extract effective text-image feature pairs in fact and sentiment perspectives and construct a fact/sentiment tension field with discrepancy metrics to capture contextual tone and polarized inconcongruities.
Outcome: The proposed method achieves state-of-the-art performance surpassing LLaVA1.5-7B with only 17.3M trainable parameters, demonstrating its optimal performance-efficiency in multi-modal sarcasm detection tasks.
A Parameter-Efficient and Fine-Grained Prompt Learning for Vision-Language Models (2025.acl-long)

Copied to clipboard

Challenge: Current vision-language models extract semantic information from large-scale cross-modal associations, limiting performance and efficiency.
Approach: They propose a detail-oriented prompt learning method to implement fine-grained multi-modal semantic alignment with merely 0.25M trainable parameters.
Outcome: The proposed method implements fine-grained multi-modal semantic alignment with merely 0.25M trainable parameters.
Multi-Scale Prompt Memory-Augmented Model for Black-Box Scenarios (2024.naacl-long)

Copied to clipboard

Challenge: Existing methods for few-shot text classification require numerous LMs’ calls to search optimal prompts, thus resulting in overfitting performance and increasing computational cost.
Approach: They propose a multi-scale knowledge prompt-based memory model that extracts instance-level and class-level knowledge and stores them in memory banks during training.
Outcome: Experiments on different benchmarks and parameter analysis demonstrate the effectiveness and efficiency of MuSKPrompt in black-box few-shot text classification tasks.
An Orthogonal High-Rank Adaptation for Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Low-rank adaptation (LoRA) efficiently adapts LLMs to downstream tasks by decomposing LLM’s weight update into trainable low-rank matrices for fine-tuning.
Approach: They propose an orthogonal high-rank adaptation for parameter-efficient fine-tuning that decomposes LLMs’ pre-trained weight matrices into orthogonals via QR decomposition and splits them into two low-redundancy high-ranked components.
Outcome: Empirical results show that OHoRA outperforms LoRA and its variants and generates task-tailored representation spaces with 0.0371% trainable parameters.

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