AI Wiki — Knowledge Base

R9V Update: Created and adopted KVA projections based on Deepseek V4.1 Flash + HySparse2/MiMo-V3 for Qwen3.8 Flash Next. This is a game changer for models that don't natively implement it. 1.45-1.85x speedup in prefill to 3k+ at a small deficit to perplexity. [2x R9700, 128GB DDR5]

Here's my *first* implementation of KVA projectors on QFN (just the uncensored model for now) the highlights are basically as follows for using the projectors at each different layer: Starting at laye

model-releasesr-localllama

24 Sept 2026

Search: “papers”

Comparative Evaluation of Static Embedding Models for HTTP Request Anomaly Detection

Compressed delayed-information projection for six-degree-of-freedom underwater vehicle navigation under delayed acoustic positioning

Confidence Falls Short: Asymmetric Certainty Gains from Optimization Hinder Multimodal Classification

Constraint-Driven Context Engineering: Designing Domain Interfaces for AI Systems

Contact-Implicit Stein Projected ADMM for Discovery of Diverse Contact-Rich Manipulation Strategies

Curriculum Learning with GNN-based Reinforcement Learning for Job Shop Scheduling

Data-driven discrete-time deep recurrent neural network-based modeling for dissipative systems

Design and Modeling of a Single-Port Three-Arm Robotic Tool for Minimally Invasive Neurosurgery

Discovery of fully efficient fault indicators along a data-based diagnosis process

EBRL: Asynchronous Embodied RL by Multi-Grained Resource Management

Enhancing Multiclass Malware Classification in Resource-Constrained Environments

Escaping Python Dependency Hell: A Hybrid Replay-and-Repair Pipeline for Python Dependency Resolution

Evaluating Open-Weight LLMs for Turkish Domain Documents Under Retrieval and Hardware Constraints

Fine-Tune, Then Rectify

Finite-Sample Probabilistic Safety Certification for AI-Based Grid-Edge Coordination

Fusion-Aware Direct 3D Gaussian Generation with Structured Patch Latent Flows

How Much Were You Told? Measuring External Information in Peer Reviews

HYDRO: Towards Non-Reversible Face De-Identification Using a High-Fidelity Hybrid Diffusion and Target-Oriented Approach

hyperbolix: Hyperbolic Deep Learning in JAX

InsurTech innovation using natural language processing

Issuer-Sovereign Agentic Payments

Latent evolving World Action Model

Learn How to Act from Your Own Interactions: On-Policy Self-Distillation for GUI Agents

Learning from Failures: Heterogeneous Graph Memory for Small Language Model Tool-Using Agents

Learning What to Activate: Combinatorial Capability Allocation for Long-Horizon Multimodal Agents

Lend me an Ear: Speech Enhancement Using a Robotic Arm with a Microphone Array

LiAM-SAM: Lifecycle-Aware Memory for Robust SAM2-Based MOT

LWCal: Loss-Weighted Calibration for Tabular Classifiers with Noisy Calibration Labels

Math Reasoning in LLMs is Organized by Approach, Not Topic

MENO: Memory-Efficient Neural Operator

MotionSpec: Spectral Trajectory Supervision for Motion-Consistent Video Generation

NADI 2026: The Second Multidialectal Arabic Speech Processing Shared Task

NaviScale: Generating Large-Scale Semantic Map Datasets for Object Navigation

NeuralSRNF: Neural Square Root Normal Fields for the Statistical Shape Analysis and Generation of Nonrigid 3D and 4D Objects

Optimizing watermarks for large language models

Parameter Importance-Driven Continual Learning for Foundation Models

PBLH Estimation from Satellite Radiances via a Dual-Encoder Transformer

Planning Trajectories that Bounce: Reflection Classes for Collision-Tolerant Robots

Pose-Aware Multimodal Automatic Tagging for Greek Traditional Music

Predicting the Progression of Adolescent Idiopathic Scoliosis

Quantization-Robust Unlearning through the Lens of Retain-Forget Loss Landscapes Interaction

R9V Update: Created and adopted KVA projections based on Deepseek V4.1 Flash + HySparse2/MiMo-V3 for Qwen3.8 Flash Next. This is a game changer for models that don't natively implement it. 1.45-1.85x speedup in prefill to 3k+ at a small deficit to perplexity. [2x R9700, 128GB DDR5]

Radiomics and artificial Intelligence for thyroid cancer diagnosis: Concepts, challenges, and solutions

Reachable Global Optimization in AI Systems: How Global Is Global?

Reflection-Aware Reasoning for Non-Line-of-Sight Pedestrian Localization

Relative Discharge Stage (RDS) Classification: A Practical Indicator of Battery Discharge Progress

Robust Trajectory Tracking of Autonomous Surface Vehicle via Lie Algebraic Online MPC

Safe learning-based control via function-based uncertainty quantification