There is a newer version of the record available.

Published 2026 | Version 0.1.1
Journal article Open

Self-Improvement Agent Harness: A Deterministic SIA Exemplar

Authors/Creators

  • 1. Active Inference Institute

Description

Abstract This exemplar documents template_sia, a deterministic implementation of the Self-Improvement Agent (SIA) harness contract described in . The default pipeline replays fixture-backed generations for the mini_classify task; opt-in live mode runs bounded target subprocesses and optional Ollama-backed meta/feedback steps. Run snapshot. Task mini_classify, run 1, 3 generation(s), live=false. Final accuracy=0.8333 over 6 held-out samples. Values are injected by scripts/z_generate_manuscript_variables.py after analysis. Keywords: self-improvement agents, benchmark harness, reproducible evaluation, agent loops --- Associated artifacts GitHub release: v0.1.1 (https://github.com/docxology/template_sia/releases/tag/v0.1.1) DOI: https://doi.org/10.5281/zenodo.20453879 Zenodo: https://zenodo.org/records/20453879 PDF SHA-256: e350973a772b52bd6971747464fa40b671a8b3a0ffb2e3a867201defc340d7f5

Files

Friedman_2026_Selfimprovement_e350973a.pdf

Files (82.6 kB)

Name Size Download all
md5:96650c6cbb82c4568c3f0189b888c51e
82.6 kB Preview Download

Additional details

Related works

Is supplement to
Software: https://github.com/docxology/template_sia/releases/tag/v0.1.1 (URL)