Copertina

An Adversarial Example for Direct Logit Attribution: Memory Management in GELU-4L

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Janiak, Jett, Rager, Can, Dao, James, Lau, Yeu-Tong
Natura:	Preprint
Pubblicazione:	2023
Soggetti:	Machine Learning Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Applying sparse autoencoders to unlearn knowledge in language models
di: Farrell, Eoin, et al.
Pubblicazione: (2024)

Chain-of-Thought Reasoning In The Wild Is Not Always Faithful
di: Arcuschin, Iván, et al.
Pubblicazione: (2025)

Automatically Finding Rule-Based Neurons in OthelloGPT
di: Singh, Aditya, et al.
Pubblicazione: (2025)

Discovering Forbidden Topics in Language Models
di: Rager, Can, et al.
Pubblicazione: (2025)

$λ$-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks
di: Pérez-Corral, Cristian, et al.
Pubblicazione: (2026)

SG-Blend: Learning an Interpolation Between Improved Swish and GELU for Robust Neural Representations
di: Sarkar, Gaurav, et al.
Pubblicazione: (2025)

MemLoss: Enhancing Adversarial Training with Recycling Adversarial Examples
di: Mahdi, Soroush, et al.
Pubblicazione: (2025)

Towards Interpretable Adversarial Examples via Sparse Adversarial Attack
di: Lin, Fudong, et al.
Pubblicazione: (2025)

Adversarial Examples Might be Avoidable: The Role of Data Concentration in Adversarial Robustness
di: Pal, Ambar, et al.
Pubblicazione: (2023)

Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization
di: Luo, Haocheng, et al.
Pubblicazione: (2026)

A New Type of Adversarial Examples
di: Nie, Xingyang, et al.
Pubblicazione: (2025)

Analyzing the Impact of Adversarial Examples on Explainable Machine Learning
di: Devabhakthini, Prathyusha, et al.
Pubblicazione: (2023)

Towards a Novel Perspective on Adversarial Examples Driven by Frequency
di: Zhang, Zhun, et al.
Pubblicazione: (2024)

Model Mimic Attack: Knowledge Distillation for Provably Transferable Adversarial Examples
di: Lukyanov, Kirill, et al.
Pubblicazione: (2024)

Generating Realistic Adversarial Examples for Business Processes using Variational Autoencoders
di: Stevens, Alexander, et al.
Pubblicazione: (2024)

Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
di: Marks, Samuel, et al.
Pubblicazione: (2024)

Logit Distance Bounds Representational Similarity
di: Nielsen, Beatrix M. G., et al.
Pubblicazione: (2026)

Challenges in Mechanistically Interpreting Model Representations
di: Golechha, Satvik, et al.
Pubblicazione: (2024)

Single Domain Generalization with Adversarial Memory
di: Yan, Hao, et al.
Pubblicazione: (2025)

Contract And Conquer: How to Provably Compute Adversarial Examples for a Black-Box Model?
di: Chistyakova, Anna, et al.
Pubblicazione: (2026)

PEAS: A Strategy for Crafting Transferable Adversarial Examples
di: Avraham, Bar, et al.
Pubblicazione: (2024)

Instant Adversarial Purification with Adversarial Consistency Distillation
di: Lei, Chun Tong, et al.
Pubblicazione: (2024)

Logit Dynamics in Softmax Policy Gradient Methods
di: Li, Yingru
Pubblicazione: (2025)

Peak-Controlled Logits Poisoning Attack in Federated Distillation
di: Tang, Yuhan, et al.
Pubblicazione: (2024)

Logit Distillation on Manifolds: Mapping by Learning
di: Yang, Yiru, et al.
Pubblicazione: (2026)

Spectral Logit Sculpting: Adaptive Low-Rank Logit Transformation for Controlled Text Generation
di: Li, Jin, et al.
Pubblicazione: (2025)

SCALA: Split Federated Learning with Concatenated Activations and Logit Adjustments
di: Yang, Jiarong, et al.
Pubblicazione: (2024)

From Projection to Prediction: Beyond Logits for Scalable Language Models
di: Dong, Jianbing, et al.
Pubblicazione: (2025)

Compute-in-Memory Implementation of State Space Models for Event Sequence Processing
di: Zhang, Xiaoyu, et al.
Pubblicazione: (2025)

Adversarial Examples in the Physical World: A Survey
di: Wang, Jiakai, et al.
Pubblicazione: (2023)

AIM: Attributing, Interpreting, Mitigating Data Unfairness
di: Liu, Zhining, et al.
Pubblicazione: (2024)

Formalising the Logit Shift Induced by LoRA: A Technical Note
di: Shi, Xiang, et al.
Pubblicazione: (2026)

Context Distillation as Latent Memory Management
di: Zheng, Ziyang, et al.
Pubblicazione: (2026)

Online Adversarial Knowledge Distillation for Graph Neural Networks
di: Wang, Can, et al.
Pubblicazione: (2021)

SHRED: Retain-Set-Free Unlearning via Self-Distillation with Logit Demotion
di: Hu, Zizhao, et al.
Pubblicazione: (2026)

Rank-Aware Spectral Bounds on Attention Logits for Stable Low-Precision Training
di: Emadi, Seyed Morteza
Pubblicazione: (2026)

Beyond Hidden-Layer Manipulation: Semantically-Aware Logit Interventions for Debiasing LLMs
di: Xia, Wei
Pubblicazione: (2025)

Model-Level GNN Explanations via Rule-to-Graph Readout for Logit Reconstruction
di: Lu, Shengyao, et al.
Pubblicazione: (2025)

Do Sparse Autoencoders Capture Concept Manifolds?
di: Bhalla, Usha, et al.
Pubblicazione: (2026)

Hierarchical Delay Attribution Classification using Unstructured Text in Train Management Systems
di: Borg, Anton, et al.
Pubblicazione: (2024)