From the 1 of 5 linked papers with an AI index.
5 papers
Composable Trust for Language Models: A proven boundary and a measured defense
Yakov Pyotr Shkolnikov
The paper proposes a trust model that places authority outside language models, using a deterministic pipeline and a non-model monitor to control operations based on source integri…
Bit-Identical Medical Deep Learning via Structured Orthogonal Initialization
Yakov Pyotr Shkolnikov
Deep learning training is non-deterministic: identical code with different random seeds produces models that agree on aggregate metrics but disagree on individual predictions, with…
Learnable Pulse Accumulation for On-Device Speech Recognition: How Much Attention Do You Need?
Yakov Pyotr Shkolnikov
Self-attention scales quadratically with sequence length, limiting transformer-based speech models on edge devices. We introduce the Learnable Pulse Accumulator (LPA), an O(n) repl…
Do Foundation Models Know Geometry? Probing Frozen Features for Continuous Physical Measurement
Yakov Pyotr Shkolnikov
Vision-language models encode continuous geometry that their text pathway fails to express: a 6,000-parameter linear probe extracts hand joint angles at 6.1 degrees MAE from frozen…
Agent Memory Below the Prompt: Persistent Q4 KV Cache for Multi-Agent LLM Inference on Edge Devices
Yakov Pyotr Shkolnikov
Multi-agent LLM systems on edge devices face a memory management problem: device RAM is too small to hold every agent's KV cache simultaneously. On Apple M4 Pro with 10.2 GB of cac…