# WMIP: A Working-Memory-Inspired Programmable Probe of Cumulative State Tracking in Large Language Models

**Authors:** Dengzhe Hou, Lingyu Jiang, Deng Li, Zirui Li, Fangzhou Lin, Kazunori D Yamada  
**Lab:** Tohoku University · **Published:** 29 Oct 2026 · **Upvotes:** 3  
**Links:** [arXiv](https://arxiv.org/abs/2603.27343) · [PDF](https://arxiv.org/pdf/2603.27343) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2603.27343.html)

## Key numbers

- **84.9%**: accuracy is 84.9% on single-step arithmetic but 31.7% on core items requiring one number after three…
- **31.7%**: accuracy is 84.9% on single-step arithmetic but 31.7% on core items requiring one number after three…
- **10%**: no model exceeds 10% accuracy

## Abstract

Human working memory lets people maintain and update information. Large language models (LLMs) face a related demand when tracking updates within a single prompt without writing out intermediate states, yet existing tests differ in their constraints on written steps. We introduce a working-memory-inspired probe of cumulative state tracking with program-computed answers, 3 to 100 updates, and final-state, intermediate-state, or multi-target queries. It compares protocols on identical items. Across 18 LLMs, accuracy is 84.9% on single-step arithmetic but 31.7% on core items requiring one number after three to seven updates. From 50 updates on, no model exceeds 10% accuracy. Core accuracy reaches 88.0% under a larger budget and instructions to write steps and place the answer on the last line, mainly due to the instruction changes rather than the budget. Core accuracy also correlates positively with public Berkeley Function Calling Leaderboard scores. These findings show that strong single-step arithmetic does not ensure reliable cumulative tracking and that output protocols substantially change measured accuracy. We release the generator, items, scoring code, and responses for reproducible comparisons across depths, query types, and protocols.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.