Papers by Yueqi Xie

5 papers
MLLM-Protector: Ensuring MLLM’s Safety without Hurting Performance (2024.emnlp-main)

Copied to clipboard

Challenge: MLLMs are deployed on limited image-text pairs, which makes them more vulnerable to catastrophic forgetting of their original abilities during safety fine-tuning.
Approach: They propose a plug-and-play strategy that detects harmful visual inputs and transforms harmful ones into harmless ones.
Outcome: The proposed approach mitigates the risks posed by malicious visual inputs without compromising the original performance of MLLMs.
Defending against Indirect Prompt Injection by Instruction Detection (2025.findings-emnlp)

Copied to clipboard

Challenge: Indirect Prompt Injection attacks can be exploited by LLMs that are embedded with external data.
Approach: They propose a detection-based approach that leverages the behavioral states of LLMs to identify potential IPI attacks.
Outcome: The proposed approach reduces the success rate of attacks to 0.03% on the BIPIA benchmark.
Red-Teaming NSFW Image Classifiers as Text-to-Image Safeguards (2026.findings-acl)

Copied to clipboard

Challenge: Not Safe for Work (NSFW) image classifiers play a critical role in safeguarding text-to-image systems.
Approach: They propose an automated red-teaming framework that leverages a set of generative AI tools to uncover NSFW image failures.
Outcome: The proposed framework uncovers and interprets failure modes and enables it to be applied to real-world T2I and T2V systems.
GradSafe: Detecting Jailbreak Prompts for LLMs via Safety-Critical Gradient Analysis (2024.acl-long)

Copied to clipboard

Challenge: Existing methods for detecting jailbreak prompts are primarily online moderation APIs or finetuned LLMs.
Approach: They propose a method which scrutinizes the gradients of safety-critical parameters in large LLMs to detect jailbreak prompts.
Outcome: The proposed method outperforms Llama Guard in detecting jailbreak prompts despite extensive finetuning with a large dataset.
Measuring Human Contribution in AI-Assisted Content Generation (2026.acl-long)

Copied to clipboard

Challenge: generative AI has created a new way to generate content with humans . varying degrees of human contribution in content generation poses significant challenges for the delineation of originality .
Approach: They propose a framework to measure human contribution in AI-assisted content generation by calculating mutual information between human input and AI-aided output relative to self-information of AI-assist output.
Outcome: The proposed measure discriminates between varying degrees of human contribution across multiple creative domains and is validated in real-world applications.

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