Aug 2026· Electronics· Vol 15, pp. 3453· 0 citations· 30 references
TL;DR
Results demonstrate that adversarial debate combined with consensus-based execution gating provides a principled framework for capital preservation across diverse market regimes.
Abstract
Existing autonomous trading systems rely on collaborative or single-agent analysis, lacking structured mechanisms for adversarial deliberation across opposing market perspectives. We propose Consensus-Gated Execution (CGX), a multi-agent architecture where specialized Bull and Bear agents engage in a three-round structured debate, with a Meta-Evaluator synthesizing their arguments to gate trade execution based on consensus strength. The system is evaluated through two complementary experiments: a 52-week aggregation study (2024) and a four-year multi-regime validation (2022–2025) across 417 biweekly sessions spanning bear, recovery, bull, and mixed market conditions. Trade signals are filtered using a tunable consensus threshold, allowing the system to balance trading frequency against signal quality. In the aggregation study, CGX achieves a Sharpe ratio of 1.90 with a maximum drawdown of 11.6%, representing a 3× improvement over trend following. In the multi-year evaluation, CGX reduces maximum drawdown by 85% and annualized volatility by 86%, with the Bear gate blocking 93% of sessions during the 2022 crash versus only 12% during the 2024 bull run. These results demonstrate that adversarial debate combined with consensus-based execution gating provides a principled framework for capital preservation across diverse market regimes.
Multi-agent systems powered by large language models (LLMs) have demonstrated potential for collaborative problem-solving, yet increasing the number of agents often introduces redundant reasoning and communication overhead, sometimes degrading performance. We propose AgentDropout, a dynamic strategy inspired by dropout regularization in neural networks, which selectively deactivates low-contribution agents during multi-agent collaboration. At each round of discussion, AgentDropout computes a semantic novelty score for every agent by measuring the divergence of its output relative to the current group consensus. Agents whose novelty score falls below an adaptive threshold are temporarily deactivated, reducing token consumption without sacrificing viewpoint diversity. We evaluate AgentDropout on mathematical reasoning (GSM8K), commonsense reasoning (StrategyQA), and collaborative code generation (HumanEval) tasks. Across three independent runs, AgentDropout achieves accuracy comparable to or modestly above fixed 5-agent debate while reducing total token consumption by 38.0–43.5% and debate rounds by 20.5% on average. Pareto analysis reveals a promising efficiency–quality trade-off, suggesting that dynamic agent deactivation may be useful for deploying multi-agent LLM systems under computational budget constraints.
Zhengxi Xiao, Qi Guo, Yuyue Wang et al.· 2026 8th International Confe...· 1 citation
A novel framework based on Koopman operator theory is developed and validates its theoretical guarantees on multi-agent consensus dynamics, making spectral certification a practical layer for trustworthy collective reasoning.
Multi-agent systems have shown strong performance in domains with reliable verifiers such as coding, where multi-parallel candidate generation selected by a verifier is effective. However, such pipelines would not generalize to open-ended, long-horizon research tasks without a verifier. While majority voting or self-consistency is often used to reach consensus as a proxy verifier, parallel agents repeatedly explore the same evidence, while access to peers'partial findings cause search to converge on an early candidate before alternatives are tested. We present ArcticSwarm, a multi-agent research architecture that separates evidence gathering from evidence integration. Subagents publish findings to a shared bulletin board, while gated isolation lets selected search tasks maintain their own prior, preventing early consensus. Structured review at three commitment boundaries enforce only confident candidates to be propagated. As a result, ArcticSwarm reaches 82.6% on the full BrowseComp-Plus set with the open-weight Qwen 3.5-27B model, compared with 78.8% without gated isolation and 74.5% additionally with structured review disabled, outperforming aligned baseline MiroFlow runs (70.6%). Extending to live-web BrowseComp, ArcticSwarm reaches 73.6% with GPT-5, which is well above the reported provider system (54.9%) and MiroFlow (63.4%). Overall, the results show that restricting peer reads during evidence gathering and strengthening commitment boundaries before a hypothesis is shared can broaden search and improve long-horizon multi-agent deep research.
Soyoung Yoon, Bo-Yi Liu, Yi-Te Wang et al.· 0 citations
ForesightSafety-TIDE, a controlled evaluation framework that strictly pairs all-honest collaboration with controlled deception by a key evidence holder and analyzes the aggregation process through multi-stage voting, testimony adoption, and evidence-root lineage propagation, reveals both the fragility of distributed fact recovery and its underlying mechanism.
Chen Yan, Zeyang Yue, Feifei Zhao et al.· 2 citations
This work introduces a collective evidence-threshold backdoor paradigm for MAS and Boundary-Conditioned Backdoor Injection, which constructs counterfactual boundary pairs to separate benign behavior before the threshold from the adversarial objective after it, and learns latent progression aligned with evidence.
Jiahao Xiao, Lei Feng, Min-Ling Zhang· 0 citations
LLM-based multi-agent trading systems, in which specialized agents collaborate through structured communication to produce trading decisions, are moving rapidly from research prototypes to live deployments that control real assets. The same inter-agent communication that makes them effective also exposes them: a corrupted signal can propagate to the final decision and translate into realized financial loss. Unlike prior attacks that presume privileged access to system internals, we restrict the adversary to what is practically reachable---the source data and prompts agents consume---yielding a low-barrier, and thus democratized threat model instantiated as role-specific adversaries. We present the first systematic empirical study in the financial domain to characterize how an adversarial signal enters a multi-agent trading system and how far it survives toward the decision. Along the role axis, we decompose a widely-used trading pipeline into four functional roles---Analyst, Researcher, Trader, and Risk Manager---and pair each with an attack matched to its interface. Along the structural axis, we evaluate four communication topologies under data- and agent-level attacks, using the Adversarial Signal Preservation Score (APS) as a post-hoc lens on why some designs are more robust than others. We conduct experiments across five assets, two backbones, and two target directions. A central finding is that no architecture is inherently robust. These findings provide insights for the future design of safer and more robust agentic trading systems.
CheolWon Na, Hao Ni, Lukasz Szpruch et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
