TextBugger: Generating Adversarial Text Against Real-world Applications (2018 · 545 citations)
The paper asks how fragile deep-learning text classifiers are, both research models and commercial cloud services, when an attacker makes small, human-readable edits to the input.
Bridging Context Gaps: Leveraging Coreference Resolution for Long Contextual Understanding (2024 · 10 citations)
LLMs still struggle with long inputs. When a document uses many pronouns, modifiers and varied noun phrases for the same entities, models lose track of the referents, especially in mid-document passages. The authors propose Long Question Coreference Adaptation (LQCA), a preprocessing pipeline that rewrites the…
Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors (2026 · 1 citation)
The paper looks at supply-chain backdoors in safety-aligned LLMs. In this setting, an attacker with write access to a checkpoint wants a hidden trigger to unlock harmful answers, while the model behaves normally otherwise. Earlier weight-editing attacks such as JailbreakEdit train the model to open with an agreeing…
“I See What You Did There”: Can Large Vision-Language Models Understand Multimodal Puns? (2026 · 1 citation)
The paper asks whether vision-language models (VLMs) actually understand multimodal puns or only react to playful-looking content.

Co-authors here

12 of 14 co-authors with a page, the most papers together first.

Venues

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 211 times (of the 149,439 papers whose bibliographies we read); 23 citing papers with a page here →.

Updated 1 November 2026