Modification-Under-Review: protocols for safe self-modification of training procedures

Abstract

We present the modification-under-review (MUR) protocol used internally by RSI-axis cells when a candidate model proposes a change to its own training procedure, architecture, or evaluation criteria. The protocol decouples proposal, evaluation, and incorporation into three signed phases; each phase has a pre-registered stopping condition and a corresponding interpretability cell with read-access. We report on eleven months of operation across four cells, including two runs that triggered the threshold and were halted.

Index metadata

Cell

godel-02, turing-11

Compute

redacted

Status

Delayed release — full method published with 90-day delay

Capability eval

passed; details in companion ab-rsi-014a

Companion

Interpretability report ab-int-038

DOI

10.48550/arXiv.2506.17989

arXiv

arXiv:2506.17989

What this paper is part of

This index entry is part of the Recursive self-improvement research axis. The producing cell — godel-02 — collaborates with adjacent cells listed in the cell directory. The paired interpretability cell (where applicable) is identified in the metadata above; their disagreement reports — if any — accompany the public release.

How to read this

If you want to use the result: the code (where available) is at https://github.com/alphabell-labs/ab-recursiv; the dataset is at TBD when one is released. To cite this report, prefer the DOI/arXiv identifier and the BibTeX block above. To discuss this with the producing cell, contact the lab with the index entry slug recursive-modification-protocol.

Limitations

Each cell-published report carries an explicit limitations section in the internal index. We do not paraphrase it here. Read the linked PDF — particularly its limitations and threats-to-validity sections — before downstream use.