🔍 Interpretability - ch1n3du

λType Theory Academic

arxiv.org·

Syracuse Views Summer 2026

🌀Complexity Science Academic

news.syr.edu·

Vision-Language Asymmetry in Bistable Image Captioning

📡Information Theory Academic

arxiv.org·

Periodic Symmetry-Adapted Encoding: Qubit Reduction in Crystalline Electronic Structure

📡Information Theory Academic

arxiv.org·

Trajectory Geometry of Transformer Representations Across Layers

💻CS Theory Academic

arxiv.org·

Market Research Deep Dive: Wrought Aluminum Turbocharger Wheels Capture 45% of Market Value – Full Size & Share Analysis by Forged vs. Cast Technology

⚙️Mechanism Design Blog

plaza.rakuten.co.jp·

Activation-Based Active Learning for In-Context Learning: Challenges and Insights

📡Information Theory Academic

arxiv.org·

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

📡Information Theory Academic

arxiv.org·

Coelho Mollo and Millière: The Vector Grounding Problem

λType Theory

philosophyofbrains.com·

Jung Kyung-Ho, Choi Soo-Young Break Up After 14 Years: Here’s Their Relationship Timeline

🕸️Network Theory News

in.mashable.com·

The technical community can't be the main character in AI safety anymore

🔗Interdisciplinary

substackcdn.com··Substack

Closure-Validated Circuit Discovery in Attention Heads: Co-activation Proposes, Ablation Disposes

🧠Cognitive Science Academic

arxiv.org·

Who Elected Anthropic?

⚖️Political Economy Blog

vizierprime.substack.com··Substack

Shared Semantics, Divergent Mechanisms: Unsupervised Feature Discovery by Aligning Semantics and Mechanisms

λType Theory Academic

arxiv.org·

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

📡Information Theory Academic

arxiv.org·

Interpreting Brain Responses to Language with Sparse Features from Language Models

🧠Cognitive Science Academic

arxiv.org·

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

🔬Philosophy of Science Academic

arxiv.org·

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

📡Information Theory Academic

arxiv.org·

BMW Highway Assistant: Over 200 million kilometres driven hands-free by customers.

Is the Space Pope Reptilian?

When Attribution Patching Lies: Diagnosis and a Second-Order Correction

Syracuse Views Summer 2026

Vision-Language Asymmetry in Bistable Image Captioning

Periodic Symmetry-Adapted Encoding: Qubit Reduction in Crystalline Electronic Structure

Trajectory Geometry of Transformer Representations Across Layers

Market Research Deep Dive: Wrought Aluminum Turbocharger Wheels Capture 45% of Market Value – Full Size & Share Analysis by Forged vs. Cast Technology

Activation-Based Active Learning for In-Context Learning: Challenges and Insights

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

Coelho Mollo and Millière: The Vector Grounding Problem

Jung Kyung-Ho, Choi Soo-Young Break Up After 14 Years: Here’s Their Relationship Timeline

The technical community can't be the main character in AI safety anymore

Closure-Validated Circuit Discovery in Attention Heads: Co-activation Proposes, Ablation Disposes

Who Elected Anthropic?

Shared Semantics, Divergent Mechanisms: Unsupervised Feature Discovery by Aligning Semantics and Mechanisms

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

Interpreting Brain Responses to Language with Sparse Features from Language Models

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

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