# Right-Sizing LLM-Agent Decomposition in VAT Determination: A Pilot Controlled Sweep

arXiv: https://arxiv.org/abs/2608.23395 · HTML: https://mlsanity.com/paper/2608.23395/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Financial trading and forecasting agents, one of 16 directions in LLM agents and tool use.
- Builds on 17 of its 32 references that are in the map.
- First posted on arXiv 24 Aug 2026.

## Placed in

- LLM agents and tool use / [Financial trading and forecasting agents](https://mlsanity.com/agents/financial-trading-and-forecasting-agents/)

## Builds on (most cited first, 17 of 17)

- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [ChatDev: Communicative Agents for Software Development](https://mlsanity.com/paper/2307.07924/) (2023)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models](https://mlsanity.com/paper/2308.11462/) (2023)
- [Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks](https://mlsanity.com/paper/2411.04468/) (2024)
- [RODE: Learning Roles to Decompose Multi-Agent Tasks](https://mlsanity.com/paper/2010.01523/) (2020)
- [A Dataset for Statutory Reasoning in Tax Law Entailment and Question Answering](https://mlsanity.com/paper/2005.05257/) (2020)
- [TDAG: A Multi-Agent Framework based on Dynamic Task Decomposition and Agent Generation](https://mlsanity.com/paper/2402.10178/) (2024)
- [When Single-Agent with Skills Replace Multi-Agent Systems and When They Fail](https://mlsanity.com/paper/2601.04748/) (2026)
- [Difficulty-Aware Agentic Orchestration for Query-Specific Multi-Agent Workflows](https://mlsanity.com/paper/2509.11079/) (2025)
- [Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline](https://mlsanity.com/paper/2601.12307/) (2026)
- [Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets](https://mlsanity.com/paper/2510.25779/) (2025)
- [Beyond the Strongest LLM: Multi-Turn Multi-Agent Orchestration vs. Single LLMs on Benchmarks](https://mlsanity.com/paper/2509.23537/) (2025)
- [Toward Expert Investment Teams:A Multi-Agent LLM System with Fine-Grained Trading Tasks](https://mlsanity.com/paper/2602.23330/) (2026)
- [An LLM Agentic Approach for Legal-Critical Software: A Case Study for Tax Prep Software](https://mlsanity.com/paper/2509.13471/) (2025)
- [Operand Quant: A Single-Agent Architecture for Autonomous Machine Learning Engineering](https://mlsanity.com/paper/2510.11694/) (2025)
- [How Does Microservice Granularity Impact Energy Consumption and Performance? A Controlled Experiment](https://mlsanity.com/paper/2502.00482/) (2025)

## Abstract (arXiv)

Recent LLM-agent systems make conflicting design bets: decompose work across many narrow agents, or use one strong tool-using agent. This pilot studies that choice on bounded cross-border VAT determination with reverse charge, where every case has an oracle label and each intermediate decision is independently scoreable. We hold the activity surface fixed (subtasks, tools, I/O schemas, validation checks, orchestrator, base model, and merge policy) and vary only the assignment of subtasks to workers across four orchestrated configurations, from one wide worker to five narrow ones, against S0, a tuned no-orchestrator single agent, with a deterministic rule engine as oracle. The program spans 4,400 runs: a 40-case, five-repeat main sweep, matched-token arms separating prompt-budget from agent-count effects, and three failure-injection arms, all judged against pre-registered falsification criteria. The two intermediate configurations lead on accuracy (0.830, against endpoints at 0.720 and 0.770) but miss the pre-stated bar against the fine endpoint, so the intermediate-optimum hypothesis remains unsupported at pilot scale. The single agent does not Pareto-dominate the orchestrated set. The matched-token criterion fires: the budget-matched single agent lands 6.5 points below the leader, but the interval includes zero, so any advantage is consistent with a prompt-budget explanation. Under injection, availability faults are absorbed at every granularity, with wide-scope restart over-recovering its baseline by +0.160, while one schema-conforming hallucinated record degrades every configuration and inverts the ordering, hitting fragmented configurations hardest. The contribution is a bounded, preregistered pilot heuristic for right-sizing decomposition (place one partition boundary at the dependency-layer midpoint), released with oracle, dataset, harness, raw traces, and analysis pipeline.