Cover Image

Representation Noising: A Defence Mechanism Against Harmful Finetuning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Rosati, Domenic, Wehner, Jan, Williams, Kai, Bartoszcze, Łukasz, Atanasov, David, Gonzales, Robie, Majumdar, Subhabrata, Maple, Carsten, Sajjad, Hassan, Rudzicz, Frank
Format:	Preprint
Published:	2024
Subjects:	Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Immunization against harmful fine-tuning attacks
by: Rosati, Domenic, et al.
Published: (2024)

Evaluating Defences against Unsafe Feedback in RLHF
by: Rosati, Domenic, et al.
Published: (2024)

Limits of Convergence-Rate Control for Open-Weight Safety
by: Rosati, Domenic, et al.
Published: (2026)

Long-form evaluation of model editing
by: Rosati, Domenic, et al.
Published: (2024)

Improving Consistency in Large Language Models through Chain of Guidance
by: Raj, Harsh, et al.
Published: (2025)

Semantic Consistency for Assuring Reliability of Large Language Models
by: Raj, Harsh, et al.
Published: (2023)

Representation Engineering for Large-Language Models: Survey and Research Challenges
by: Bartoszcze, Lukasz, et al.
Published: (2025)

Consistency in Language Models: Current Landscape, Challenges, and Future Directions
by: Novikova, Jekaterina, et al.
Published: (2025)

Resolving Lexical Bias in Model Editing
by: Rizwan, Hammad, et al.
Published: (2024)

Dependency Parsing is More Parameter-Efficient with Normalization
by: Gajo, Paolo, et al.
Published: (2025)

LLMs Underperform Graph-Based Parsers on Supervised Relation Extraction for Complex Graphs
by: Gajo, Paolo, et al.
Published: (2026)

A Game-Theoretic Approach for PMU Deployment Against False Data Injection Attacks
by: Maleki, Sajjad, et al.
Published: (2024)

Fundamental Limitations in Pointwise Defences of LLM Finetuning APIs
by: Davies, Xander, et al.
Published: (2025)

Embedding-based classifiers can detect prompt injection attacks
by: Ayub, Md. Ahsan, et al.
Published: (2024)

Voluntary Collusion with Secret Tools in Competing LLM Agents
by: Zeng, Xijie, et al.
Published: (2026)

Individualised Counterfactual Examples Using Conformal Prediction Intervals
by: Adams, James M., et al.
Published: (2025)

Testing Realism in Quantum Mechanics Through Charge Conservation
by: Atanasov, Victor
Published: (2025)

Red Teaming AI Red Teaming
by: Majumdar, Subhabrata, et al.
Published: (2025)

Quantum Advantage for Coordinated Frequency Selection Against Distributed Jammers
by: Wehner, Stephanie
Published: (2026)

Elicitor Specific Mechanisms of Defence Priming in Oak Seedlings Against Powdery Mildew
by: Rosa Sanchez‐Lucas, et al.
Published: (2025)

Single-Configuration Attack Success Rate Is Not Enough: Jailbreak Evaluations Should Report Distributional Attack Success
by: Maple, Carsten, et al.
Published: (2026)

Faithful or Fabricated? A Causal Framework for Rationalization Bias in LLM Judges
by: Tapwal, Riya, et al.
Published: (2026)

PRISM: Generation-Time Detection and Mitigation of Secret Leakage in Multi-Agent LLM Pipelines
by: Tapwal, Riya, et al.
Published: (2026)

Justified Evidence Collection for Argument-based AI Fairness Assurance
by: Sabuncuoglu, Alpay, et al.
Published: (2025)

Towards Robust Federated Analytics via Differentially Private Measurements of Statistical Heterogeneity
by: Scott, Mary, et al.
Published: (2024)

Private Federated Multiclass Post-hoc Calibration
by: Maddock, Samuel, et al.
Published: (2025)

FLAIM: AIM-based Synthetic Data Generation in the Federated Setting
by: Maddock, Samuel, et al.
Published: (2023)

DriveSafe: A Hierarchical Risk Taxonomy for Safety-Critical LLM-Based Driving Assistants
by: Kumar, Abhishek, et al.
Published: (2026)

Wavelet Representation and Sampling of Complex Luminaires
by: A. Atanasov, et al.
Published: (2025)

CAFe: Unifying Representation and Generation with Contrastive-Autoregressive Finetuning
by: Yu, Hao, et al.
Published: (2025)

Exploring the features used for summary evaluation by Human and GPT
by: Sadeghi, Zahra, et al.
Published: (2025)

Scenarios and Approaches for Situated Natural Language Explanations
by: Qiu, Pengshuo, et al.
Published: (2024)

Understanding Language Model Circuits through Knowledge Editing
by: Ge, Huaizhi, et al.
Published: (2024)

How Well Can Knowledge Edit Methods Edit Perplexing Knowledge?
by: Ge, Huaizhi, et al.
Published: (2024)

SoftAdaClip: A Smooth Clipping Strategy for Fair and Private Model Training
by: Soleymani, Dorsa, et al.
Published: (2025)

On the Limitations of Speaker Diarization
by: Joana Amorim, et al.
Published: (2026)

Representational Harms in LLM-Generated Narratives Against Global Majority Nationalities
by: Nguyen, Ilana, et al.
Published: (2026)

Nachruf: Hans‐Herbert Schmidtke
by: Mihail Atanasov, et al.
Published: (2024)

Threat, Risk and Mitigation Taxonomy for Digital Identity Systems
by: SHEIK, AL TARIQ, et al.
Published: (2024)

Unveiling the Interplay: Salinity‐Modulated Defence Mechanisms in Medicago truncatula Against Phoma medicaginis Infection
by: Manel Chaouachi, et al.
Published: (2025)