Latent Trajectory Surgery: Editing Agent Plans in Mid-Run

Abstract

When a paired interpretability cell wants to test whether a particular sub-plan is load-bearing for an agent's behaviour, the cleanest experiment is to surgically remove or replace that sub-plan and re-execute. We make this practical: a substrate-integrated surgery tool that can edit a substrate-hosted agent's plan at a specific causal joint, while preserving the upstream state. We demonstrate the tool's use on three case-study halts called by paired interpretability cells in 2024.

Index metadata

Cell

cantor-18

Compute

16 H100-days

Status

Open release

Code

github.com/alphabell-labs/ab-surgery

DOI

10.48550/arXiv.2412.04881

arXiv

arXiv:2412.04881

What this paper is part of

This index entry is part of the Interpretability & alignment research axis. The producing cell — cantor-18 — 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-surgery; 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 latent-trajectory-surgery.

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.