AI Wiki — Knowledge Base

I turned Qwen3.8-27B Q2_64 + llama.cpp into a fully TypeSafe AI-compatible Jev-like system. OpenAI API still intact! World’s first Vision-enabled Jev-like model! <10 GB VRAM, 170 ms on an RTX 3090 and ~140 tok/s in chat. 76% vs. 88% Jev-1.13 Acc. on a diverse 22,000-request typed-decision benchmark

Yesterday I released Bonsai-Llama-Jev - a Jev-like Typed Decision inference system for locally operated, sovereign AI. In my new typed-decision-bench, covering more than 22,000 decisions across 175 us

model-releasesr-localllama

23 Sept 2026

Search: “deployment”

SlackDrive: Reclaiming Runtime Slack for Adaptive Driving Inference

Text Scores Can Miss Waveform Use: A Qwen2-Audio Quantization Case Study

The Like Trap: Multi-Stage Poisoning against Agents in Similarity-based Recommendation Systems

UniDataAgent: An Ontology-Grounded Agent for Enterprise Question-to-Report Automation

Validation and Simulation Catch Different Errors: Four Levels of Evaluation for LLM-Generated Circuits

What Confidence Routing Is Actually Doing: Auditing Routing, Calibration, and Commitment in Multi-Agent Deliberation

'What's That Sound?': A Versatile, Robust, and Lightweight Convolutional Transformer for Environment Sound Recognition

When Accuracy Gaps Fail to Certify: Auditing Cross-Domain Recalibration of LLM Judges

When Adaptation Hurts: Split Sensitivity and Person-Level Negative Transfer in Federated Wearable Onboarding

When Clients Are Orchestrated: Strategic Gradient Manipulation to Defeat Federated Learning Servers with Efficient Defense

Annual Earth-observation embeddings encode wildfire disturbance and support simplified burned area mapping

'As a Language Model...': Chat Template Switches LLM Self-Referential Voice and Activation Steering Reproduces It

BDSLI: A hybrid CNN-Transformer model for Bengali Sign Language interpretation

Beyond Raw Engagement: A Counterfactual Observability Framework for Recommender Systems at Netflix

CableVLA: Simulation-Privileged Global-Local Representation Learning for Cable Routing

Calibration as a First-Class Criterion in LLM Evaluation

CompKV: Compensation-Aware KV Selection for Long-Context LLM Inference

Controlling risks of AI in chemical science with agents

Diffusion Drafts, AR Verifies: Accelerating Document OCR with Self-Speculative Decoding

DUMA-Bench: A Dual-Control Multi-Agent Benchmark for Evaluating LLM Agent Security

DVA-Neurons: Design and Verification of Adaptive LIF Neurons: From Single-Neuron Dynamics to Multi-Neuron Spiking Networks

Efficient Iterative Retrieval with Heterogeneous Batching

ENPIRE: Agentic Robot Policy Self-Improvement in the Real World

Evaluating Accuracy and Probabilistic Reliability of Zero-Shot Time Series Foundation Models

Explainable Recommendations at Scale: LLM Rationales for YouTube Music Artist Discovery

Federating Quantum and Classical Computing: A Privacy-Preserving Hybrid Approach

Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs

From Code to Requirements: Agentic Reverse Engineering of Business Rules at Enterprise Scale

From IceCube to IT-Sphere: A Hybrid Quantum-Classical GNN for Banking IT Root Cause Analysis

From Research Frontier to Laboratory Bench: Design of a Four-Tier Experimental Teaching System for Multimodal Medical Image Intelligent Diagnosis

Generalizing Manipulation Skills with a Local Coding Agent

GTR: Gated Token Recurrence for Efficient Dense Prediction

Hierarchical Floorplan-Guided Vision-Language Exploration for Embodied Question Answering

How Many Pixels Is a Digit Worth? Place-Aware Coordinate Entropy for GUI Agent Confidence Estimation

Impact Is Not Invalidation: Ask About the Claim, Not the Diff

IndustrialVLA-Bench: A Traceable Multi-Axis Evaluation of Open Robot Policy Models

Leaky-integrator reconstruction: taming error accumulation in recursive differenced time-series forecasting

Leveraging Vision-Based Point Cloud Map Priors for Camera-Based 3D Object Detection and Online Vectorized HD Mapping

LiDAS: Lighting-driven Dynamic Active Sensing for Nighttime Perception

Lifelong Learning of Video Diffusion Models From a Single Video Stream

Linear probing enables Ship-Radiated Noise recognition with pretrained audio embeddings

LoRango: It Takes Two LoRAs to Unlock Hidden Behaviors in Diffusion Models

Manage Kubernetes Node Fleets with NodeWright

MedVLA: A Hierarchical Vision-Language-Action Framework for Closed-Loop Precision Medical Robot Manipulation

MicroPush: A Lightweight Simulator for Sim-to-Real Microrobotic Pushing and Assembly

Multi-Term Fourier Graph Neural Network with Sample Relationship Learning for Enhanced Remaining Useful Life Prediction

NostrAgent: A Decentralized Identity and Delegation Architecture for Sovereign Agentic Systems