---
title: "Achiral Papers"
description: "Achiral research preprints on AI memory, activation-guided graph retrieval, and Jacobian-causal memory control."
reviewed: "2026-08-24"
---

# Achiral Papers

Canonical HTML: https://achiral.ai/papers

Notes, papers, and experiments toward Emergent Memory Systems.

## Jacobian-Causal Memory Control for Persistent Memory in Language Model Systems
Authors: Marvin Danig
Venue: Architecture and analysis preprint
Date: 2026-07-30
Canonical page: https://achiral.ai/papers/jacobian-causal-memory-control-for-persistent-memory-in-language-model-systems
PDF: https://achiral.ai/papers/jacobian-causal-memory-control-for-persistent-memory-in-language-model-systems.pdf
Research essay: https://achiral.ai/research/jacobian-causal-memory-control
Abstract:
Persistent-memory language models do more than recall the past. They decide which past experiences get to steer present behavior. Most memory systems rank records by relevance, recency, summaries, or tool state. But they do not ask the harder question: what will this memory make the model do next? We treat recall as an inference-time intervention. Each memory carries an influence signature: an estimate of how it may shift future model behavior when it enters the workspace through context, memory tools, attention, or model state. A controller admits a memory only when its expected effect, side effects, policy risk, and reliability fit the task. It then observes what happened, records lineage, and rolls back or quarantines memories whose effects diverge from prediction. Trust in a memory then rests on its observed effect, not on its relevance.
Boundary: Research boundary: this is an architecture-and-analysis preprint, not a benchmark report. It defines claim boundaries, falsification tests, evaluation protocols, and deployment risks for future measurement.
Topics: AI memory, Jacobian-causal memory control, inference-time control, agent memory

## Activation-Guided Graph Retrieval for Cognitive Memory Reconstruction in Language Model Systems
Authors: Marvin Danig
Venue: Architecture and analysis preprint
Date: 2026-07-20
Canonical page: https://achiral.ai/papers/activation-guided-graph-retrieval-for-cognitive-memory-reconstruction-in-language-model-systems
PDF: https://achiral.ai/papers/activation-guided-graph-retrieval-for-cognitive-memory-reconstruction-in-language-model-systems.pdf
Research essay: https://achiral.ai/research/activation-guided-graph-retrieval
Abstract:
Most retrieval-augmented language models choose context by query relevance. Long-lived AI systems need another signal: whether a memory is available for the current goal. We describe an activation-guided graph retrieval architecture. Request and goal cues start traversal inside an authorized memory graph. Activation and fan effects keep the search bounded. Retrieved entities emit new cues, and the system rebuilds a typed evidence graph before generation. The proposal gives five testable claims about candidate selection, traversal control, graph regularization, evidence reconstruction, and reinforcement. It also defines cost estimates, ablations, failure cases, and a reproducible evaluation protocol.
Boundary: Research boundary: this is an architecture-and-analysis preprint, not a benchmark report. It defines claim boundaries, falsification tests, evaluation protocols, and deployment risks for future measurement.
Topics: AI memory, activation-guided graph retrieval, ACT-R, evidence graphs