🔍 AI Interpretability - jimman · Scour

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

✍️Prompt Engineering Academic

Inside the Visual Mind: Neuroscience-Motivated Concept Circuits for Interpreting and Steering Vision Transformers

✍️Prompt Engineering Academic

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

⚡Model Efficiency Academic

TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment

⚡LLM Optimization Academic

A Unifying Framework for Concept-Based Representational Similarity

⚡LLM Optimization Academic

Temporal Preference Concepts and their Functions in a Large Language Model

✍️Prompt Engineering Academic

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

⚡LLM Optimization Academic

Inside the LLM Word Factory

⚡LLM Optimization Academic

The Rival Theologies of Artificial Intelligence

✍️Prompt Engineering

palladiummag.com·

Vision-Language Asymmetry in Bistable Image Captioning

✍️Prompt Engineering Academic

Priors Persist Through Suppression: A Stroop Paradigm for Lexical Override

✍️Prompt Engineering Academic

Whisper Hallucination Detection and Mitigation via Hidden Representation Steering and Sparse AutoEncoders

🤖AI Academic

RGB-S: Image-Aligned Tactile Saliency for Robust Dexterous Manipulation

✍️Prompt Engineering Academic

The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models

⚡LLM Optimization Academic

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

⚡LLM Optimization Academic

DiffoR: A Unified Continuous Generative Framework for Universal Ordinal Regression

⚡LLM Optimization Academic

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

✍️Prompt Engineering Academic

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

🤖AI Academic

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

✍️Prompt Engineering Academic

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

📡RSS Academic

Sign up or log in to see more results

Log in to enable infinite scrolling