🔍 AI Interpretability - justjcullen · Scour

Shared Latent Structures Enable Unified Backdoor Detection and Mitigation in LLMs

🔢Homomorphic Encryption Academic

Ablation-Reversible Heads Don't Transfer: A Stress Test for Mechanistic Role Claims in Transformers

⚡Incremental Computation Academic

Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamics

🤖Software Engineering, AI, Personal Knowledge Mangement, Strongly Typed Languages, Math, Abstractions, Data Models, Event Sourcing Academic

arxiv.org··Cited by 1 article

Set-Based Transformer for Atmospheric Compensation in Standoff LWIR Hyperspectral Imaging

λFunctional Programming Academic

A Unifying Framework for Concept-Based Representational Similarity

🔢Homomorphic Encryption Academic

Measuring a hate speech spectrum with faceted Rasch item response theory and perspective-aware, explainable-by-design deep learning

🔗Dependent Types Academic

Inside the LLM Word Factory

🤖Software Engineering, AI, Personal Knowledge Mangement, Strongly Typed Languages, Math, Abstractions, Data Models, Event Sourcing Academic

Priors Persist Through Suppression: A Stroop Paradigm for Lexical Override

⚡Incremental Computation Academic

Interpreting Brain Responses to Language with Sparse Features from Language Models

∘Category Theory Academic

The Amplifying Mirror: Locating and Steering the Partisan Direction inside a Large Language Model

⚡Incremental Computation Academic

DiffoR: A Unified Continuous Generative Framework for Universal Ordinal Regression

⚡Incremental Computation Academic

When Built-in Thinking Helps and Hurts: Constraint-Level Error Shifts in Instruction Following

🔗Dependent Types Academic

The Neutral Mask: How RLHF Provides Shallow Alignment while Leaving Partisan Structure Intact in a Large Language Model

🎯Reinforcement Learning Academic

Wearable Single-Lead ECG Detects Fine-Grained Structural Heart Disease Through Echo-Report Supervision

👁️Observability Academic

VFUSE: Virulent Feature Understanding with Sparse autoEncoders

🔢Homomorphic Encryption Academic

Log in to enable infinite scrolling