AI Wiki — Knowledge Base

日経デジタルガバナンスにて、Sakana AIの「事後学習」技術と Sakana Chat / Namazuモデルに関する記事が掲載されました。 Namazuモデル開発をリードしたリサーチサイエンティストの菅沼雅徳とチーフサイエンティストの秋葉拓哉に取材いただきました。記事: …

日経デジタルガバナンスにて、Sakana AIの「事後学習」技術と Sakana Chat / Namazuモデルに関する記事が掲載されました。 Namazuモデル開発をリードしたリサーチサイエンティストの菅沼雅徳とチーフサイエンティストの秋葉拓哉に取材いただきました。記事: https://www.nikkei.com/prime/digital-governance/article/DGXZ

researchdavid-ha--x

7 May 2026

Search: “research”

Multiscale Euclidean Network Trajectories: Second-Moment Geometry, Attribution, and Change Points

Neural Discovery of Strichartz Extremizers

Neural-Guided Domain Restriction to Accelerate Pseudospectra Computation for Structured Non-normal Banded Matrices

Norm Anchors Make Model Edits Last

Not Every Subject Should Stay: Machine Unlearning for Noisy Engagement Recognition

NSL-MT: Linguistically Informed Negative Samples for Efficient Machine Translation in Low-Resource Languages

On the Architectural Complexity of Neural Networks

On the Influence of the Feature Computation Budget on Per-Instance Algorithm Selection for Black-Box Optimization

On the Non-decoupling of Supervised Fine-tuning and Reinforcement Learning in Post-training

On the Wasserstein Gradient Flow Interpretation of Drifting Models

Online Nonstochastic Prediction: Logarithmic Regret via Predictive Online Least Squares

Optimize-at-Capture: Highly-adaptive Exposure Controlling for In-Vehicle Non-contact Heart-rate Monitoring

Pack it in: Packing into Partially Filled Containers Through Contact

PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data

PhysForge: Generating Physics-Grounded 3D Assets for Interactive Virtual World

Physics-Guided Regime Unmixing

Physiologically Grounded Driver Behavior Classification: SHAP-Driven Elite Feature Selection and Hybrid Gradient Boosting for Multimodal Physiological Signals

Position: the Stochastic Parrot in the Coal Mine. Model Collapse is a Threat to Low-Resource Communities

PRISM: Color-Stratified Point Cloud Sampling

Probabilistic Classification and Uncertainty Quantification of Sahara Desert Climate Using Feedforward Neural Networks

Programmatic Context Augmentation for LLM-based Symbolic Regression

Progressive J-Invariant Self-supervised Learning for Low-Dose CT Denoising

QuadBox: Accelerating 3D Gaussian Splatting with Geometry-Aware Boxes

QUIVER: Cost-Aware Adaptive Preference Querying in Surrogate-Assisted Evolutionary Multi-Objective Optimization

RaguTeam at SemEval-2026 Task 8: Meno and Friends in a Judge-Orchestrated LLM Ensemble for Faithful Multi-Turn Response Generation

RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent

Reduced-order Neural Modeling with Differentiable Simulation for High-Detail Tactile Perception

Replay-Based Continual Learning for Physics-Informed Neural Operators

RetentiveKV: State-Space Memory for Uncertainty-Aware Multimodal KV Cache Eviction

Reward-Guided Semantic Evolution for Test-time Adaptive Object Detection

Right Model, Right Time: Real-Time Cascaded-Fidelity MPC for Bipedal Walking

ROCm Status in mid 2026 [D]

Row-stochastic matrices can provably outperform doubly stochastic matrices in decentralized learning

Same Voice, Different Lab: On the Homogenization of Frontier LLM Personalities

SAMIC: A Lightweight Semantic-Aware Mamba for Efficient Perceptual Image Compression

SCOUT: Active Information Foraging for Long-Text Understanding with Decoupled Epistemic States

ScriptHOI: Learning Scripted State Transitions for Open-Vocabulary Human-Object Interaction Detection

Segmenting proto-halos with vision transformers

Sentiment Analysis and Customer Satisfaction Prediction on E-Commerce Platforms Based on YouTube Comments Using the XGBoost Algorithm

Simultaneous CNN Approximation on Manifolds with Applications to Boundary Value Problems

Sparsity is Combinatorial Depth: Quantifying MoE Expressivity via Tropical Geometry

Stop Automating Peer Review Without Rigorous Evaluation

Structured 3D Latents Are Surprisingly Powerful: Unleashing Generalizable Style with 2D Diffusion

Superposition Is Not Necessary: A Mechanistic Interpretability Analysis of Transformer Representations for Time Series Forecasting

Taming Outlier Tokens in Diffusion Transformers

The Download: the tech reshaping IVF and the rise of balcony solar

The First Token Knows: Single-Decode Confidence for Hallucination Detection