AI Wiki — Knowledge Base

VLMは人間のような創造性を持てるか？ケネス・スタンレー教授らの『目標という幻想（Why Greatness Cannot Be Planned）』は、明確な目標を設定することが、かえって真に偉大な発見を遠ざけてしまうという逆説を論じた書籍です。その議論の中核にあったのが「Pi…

VLMは人間のような創造性を持てるか？ケネス・スタンレー教授らの『目標という幻想（Why Greatness Cannot Be Planned）』は、明確な目標を設定することが、かえって真に偉大な発見を遠ざけてしまうという逆説を論じた書籍です。その議論の中核にあったのが「PicBreeder」の実験でした。 PicBreeder では、ユーザーが「面白い」と感じた画像を選び、それを少しずつ進化

researchdavid-ha--x

11 Jul 2026

Search: “research”

Same Loss, Same Noise, Opposite Schedules: Noise Structure and Optimizer Normalization Jointly Determine Whether Learning-Rate Cooldown Helps

Saturation Makes Quantization Error Additive: A Coverage Model with a Certificate

Segregate, Refine, Integrate: Decomposing Multimodal Fusion for Sentiment Analysis

Semantic-Edge Response Decoding of SAM3 for Zero-Shot Crack Segmentation

Semidirect Fourier Delta Attention: Phase-Controlled Delta Memory with Constructive Chunk-WY Kernels

SinAE: A Single-Architecture Flow-Matching Autoencoder for Cross-Domain Atomic Systems

SlimPer: Make Personalization Model Slim and Smart

Solution of the Hempel's statistical ambiguity problem and Causal AI

Some Complexity Results for Robustness Verification for Binarized Neural Networks

Sparse Inter-Layer Dependencies of Transformer FFN Neurons

Spectral Diffusion Processes

SpikeDS: Dual Sparsity Spikformer for Perineural Invasion Prediction in 3D MRI

Statistical Non-linear Reconstruction Loss for Image Anomaly Detection

Steering Diffusion Models via Class-Contrastive Influence for Few-Shot Medical Classification

TAKE: Trajectory-Aware Knowledge Estimation for Text Dataset Distillation

The Computational Basis of Confidence in Large Language Models

The Geometry of Memorization: Finite-Time Spectral Sensitivity as a Diagnostic for Flow Matching Models

The Limits and Potentials of Local SGD for Distributed Heterogeneous Learning with Intermittent Communication

The Seriality Gap in Video Diffusion Models

The Spectrum Is Not Enough: When Context Helps Time-Series Forecasting

Toward Metaphor-Fluid Conversation Design for Voice User Interfaces

Towards Vision-Free CIR: Attribute-Augmented Scoring and LLM-Based Reranking for Zero-Shot Composed Image Retrieval

Traceback Translators Against Forgetting in Continual Fake Speech Detection

Translation as a Computationally Efficient Bridge: Feasibility of English BERT for Low-Resource Languages

Uncertainty-Aware Multi-Source Retinal Fluid Segmentation in OCT

Understanding Structured Health Data through Interaction-Aware Mixture-of-Experts

UniMedSeg: Unified In-Context Learning for Multi-Paradigm 2D/3D Medical Image Segmentation

Virtual Chromoendscopy with Tunable Visibility Enhancement

Visual Access Boundaries in Vision-Language Model Reasoning

Visual Species Recognition with Large Multimodal Models as Post-Hoc Correctors

We scaled a robot model natively to 8,000 timesteps of context, 5 minutes worth of muscle memory, with constant inference cost. Robot polici…

Weakly Supervised Spatio-Temporal Candidate Discovery of Dairy Farm Sites from Seasonal Satellite Imagery

1) If you haven't read AI as Normal Technology, these annotated slides are probably the easiest way to get a high-level overview. https://ww…

Google named a Leader in the 2026 IDC MarketScape for Worldwide Foundation Model Software

🤖 Un robot français en seulement 9 mois ! Rémi Cadene, ancien de Tesla AI et Hugging Face, explique comment son équipe veut faire des robot…

Everyone keeps asking me how to build a second brain or an LLM wiki. Here is the easiest setup I have found. I took my Wiki Builder skill, i…

Language models and coding agents are great, but there is more to life, and more to AI, than just LLM agents.

A Graph Neural Network Model for Real-Time Gesture Recognition Based on sEMG Signals

A New Human-Likeness and Comfort Index for Robot Movements Along Prescribed Paths

A Quantum Reservoir Architecture for Chaotic Forecasting and a Test of Whether Its High Dimension Helps

A Self-Supervised Approach for Minimal-Annotation Hydroacoustic Data Exploration

Adversarial Decoys: Misdirecting Attention-Based Defenses in ViT

Adversarial Social Epistemology for Assemblies of Humans and Large Language Models

AI-guided stimuli discovery and generation to optimize facial emotion perception studies in autism

AI-integrated models for assessing agricultural resilience

Also, is Ultra better than Pro or is it the same thing but different nomenclature?

An interpretable Good--Turing restart criterion for k-means++