Xin Xu

XinXuNLPer

·

https://xxupiano.github.io/

AI & ML interests

Mechanistic Interpretability of Trustworthy LLMs and Music LLMs

Recent Activity

upvoted a paper 1 day ago

Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?

upvoted a paper 6 days ago

MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use

authored a paper 6 days ago

Evaluating Music Context Preservation: A Multi-facet Framework for Music Editing Systems

View all activity

Organizations

commented 2 papers 11 months ago

When Benchmarks Age: Temporal Misalignment through Large Language Model Factuality Evaluation

Paper • 2510.07238 • Published Oct 8, 2025 • 15 •

BiasFreeBench: a Benchmark for Mitigating Bias in Large Language Model Responses

Paper • 2510.00232 • Published Sep 30, 2025 • 16 •

commented a paper over 1 year ago

BiasEdit: Debiasing Stereotyped Language Models via Model Editing

Paper • 2503.08588 • Published Mar 11, 2025 • 7 •