Jade (Lei) Yu
  • Bio
  • Papers
  • News
  • Experience
  • Recent & Upcoming Talks
    • Example Talk
  • Publications
    • Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations
    • Infinite Mixture Chaining: An Efficiency-Based Framework for the Dynamic Construction of Word Meaning
    • The Reasoning-Memorization Interplay in Language Models Is Mediated by a Single Direction
    • Beyond Sparse Rewards: Enhancing Reinforcement Learning with Language Model Critique in Text Generation
    • Drlc: Reinforcement learning with dense rewards from llm critic
    • Emergence of a high-dimensional abstraction phase in language transformers
    • Enhancing reinforcement learning with dense rewards from language model critic
    • Functional faithfulness in the wild: Circuit discovery with differentiable computation graph pruning
    • Geometric Signatures of Compositionality Across a Language Model's Lifetime
    • Geometric Signatures of Compositionality in Language Models
    • Intrinsic evaluation of unlearning using parametric knowledge traces
    • Mechanisms of non-factual hallucinations in language models
    • Mechanistic understanding and mitigation of language model non-factual hallucinations
    • Robust LLM safeguarding via refusal feature adversarial training
    • A computational analysis of crosslinguistic regularity in semantic change
    • Rolecraft-glm: Advancing personalized role-playing in large language models
    • Systematic word meta-sense extension
    • Word sense extension
    • Infinite mixture chaining: Efficient temporal construction of word meaning
    • Noun2Verb: Probabilistic frame semantics for word class conversion
    • Predicting emergent linguistic compositions through time: Syntactic frame extension via multimodal chaining
    • How nouns surface as verbs: Inference and generation in word class conversion
    • Inferring symmetry in natural language
    • An example preprint / working paper
    • An example journal article
    • An example conference paper
  • Projects
  • Blog
    • 🎉 Easily create your own simple yet highly customizable blog
    • 🧠 Sharpen your thinking with a second brain
    • 📈 Communicate your results effectively with the best data visualizations
    • 👩🏼‍🏫 Teach academic courses
    • ✅ Manage your projects
  • Projects
    • Pandas
    • PyTorch
    • scikit-learn
  • Teaching
    • Learn JavaScript
    • Learn Python
  • Work Experience

Robust LLM safeguarding via refusal feature adversarial training

Jan 1, 2024·
Lei Yu
,
Virginie Do
,
Karen Hambardzumyan
,
Nicola Cancedda
· 0 min read
Cite
Type
Journal article
Publication
arXiv preprint arXiv:2409.20089
Last updated on Jan 1, 2024
← Mechanistic understanding and mitigation of language model non-factual hallucinations Jan 1, 2024
A computational analysis of crosslinguistic regularity in semantic change Jan 1, 2023 →

© 2025 Me. This work is licensed under CC BY NC ND 4.0

Published with Hugo Blox Builder — the free, open source website builder that empowers creators.