SafeSieve: From Heuristics to Experience in Progressive Pruning for LLM-based Multi-Agent Communication

• URL: http://arxiv.org/abs/2508.11733v1
• Date: Fri, 15 Aug 2025 13:44:50 GMT
• Title: SafeSieve: From Heuristics to Experience in Progressive Pruning for LLM-based Multi-Agent Communication
• Authors: Ruijia Zhang, Xinyan Zhao, Ruixiang Wang, Sigen Chen, Guibin Zhang, An Zhang, Kun Wang, Qingsong Wen,
• Abstract summary: We present SafeSieve, a progressive and adaptive multi-agent pruning algorithm.<n>We show that SafeSieve achieves 94.01% average accuracy while reducing token usage by 12.4%-27.8%.<n>These results establish SafeSieve as a robust, efficient, and scalable framework for practical multi-agent systems.
• Score: 19.633176635669397
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: LLM-based multi-agent systems exhibit strong collaborative capabilities but often suffer from redundant communication and excessive token overhead. Existing methods typically enhance efficiency through pretrained GNNs or greedy algorithms, but often isolate pre- and post-task optimization, lacking a unified strategy. To this end, we present SafeSieve, a progressive and adaptive multi-agent pruning algorithm that dynamically refines the inter-agent communication through a novel dual-mechanism. SafeSieve integrates initial LLM-based semantic evaluation with accumulated performance feedback, enabling a smooth transition from heuristic initialization to experience-driven refinement. Unlike existing greedy Top-k pruning methods, SafeSieve employs 0-extension clustering to preserve structurally coherent agent groups while eliminating ineffective links. Experiments across benchmarks (SVAMP, HumanEval, etc.) showcase that SafeSieve achieves 94.01% average accuracy while reducing token usage by 12.4%-27.8%. Results further demonstrate robustness under prompt injection attacks (1.23% average accuracy drop). In heterogeneous settings, SafeSieve reduces deployment costs by 13.3% while maintaining performance. These results establish SafeSieve as a robust, efficient, and scalable framework for practical multi-agent systems. Our code can be found in https://anonymous.4open.science/r/SafeSieve-D8F2FFUN.

Related papers

• Agentic Test-Time Scaling for WebAgents [65.5178428849495]
  We present Confidence-Aware Test-Time Scaling (CATTS), which uses vote-derived uncertainty to allocate compute only when decisions are genuinely contentious.<n>CATTS improves performance on WebArena-Lite and GoBrowse by up to 9.1% over React while using up to 2.3x fewer tokens than uniform scaling.
  arXiv  Detail & Related papers  (2026-02-12T18:58:30Z)
• Zero-Shot Embedding Drift Detection: A Lightweight Defense Against Prompt Injections in LLMs [2.2448294058653455]
  adversarial prompts exploit indirect input channels such as emails or user-generated content to circumvent alignment safeguards.<n>We propose Zero-Shot Embedding Drift Detection (ZEDD), a lightweight, low-engineering-overhead framework that identifies both direct and indirect prompt injection attempts.<n>ZEDD operates without requiring access to model internals, prior knowledge of attack types, or task-specific retraining.
  arXiv  Detail & Related papers  (2026-01-18T11:33:35Z)
• ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack [52.17935054046577]
  We present ReasAlign, a model-level solution to improve safety alignment against indirect prompt injection attacks.<n>ReasAlign incorporates structured reasoning steps to analyze user queries, detect conflicting instructions, and preserve the continuity of the user's intended tasks.
  arXiv  Detail & Related papers  (2026-01-15T08:23:38Z)
• Bilevel Optimization for Covert Memory Tampering in Heterogeneous Multi-Agent Architectures (XAMT) [0.0]
  Multi-Agent Systems (MAS) are inherently heterogeneous, integrating conventional Multi-Agent Reinforcement Learning (MARL) with emerging Large Language Model (LLM) agent architectures.<n>A critical shared vulnerability is reliance on centralized memory components: the shared Experience Replay (ER) buffer in MARL and the external Knowledge Base (K) in RAG agents.<n>This paper proposes XAMT (Bilevel Optimization for Covert Memory Tampering in Heterogeneous Multi-Agent Architectures), a novel framework that formalizes attack generation as a bilevel optimization problem.
  arXiv  Detail & Related papers  (2025-12-15T23:04:48Z)
• AgentShield: Make MAS more secure and efficient [5.105635962432747]
  AgentShield is a distributed framework for efficient, decentralized auditing.<n>AgentShield achieves a 92.5% recovery rate and reduces auditing overhead by over 70% compared to existing methods.
  arXiv  Detail & Related papers  (2025-11-28T06:55:50Z)
• AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning [78.5751183537704]
  AdvEvo-MARL is a co-evolutionary multi-agent reinforcement learning framework that internalizes safety into task agents.<n>Rather than relying on external guards, AdvEvo-MARL jointly optimize attackers and defenders.
  arXiv  Detail & Related papers  (2025-10-02T02:06:30Z)
• Eigen-1: Adaptive Multi-Agent Refinement with Monitor-Based RAG for Scientific Reasoning [53.45095336430027]
  We develop a unified framework that combines implicit retrieval and structured collaboration.<n>On Humanity's Last Exam (HLE) Bio/Chem Gold, our framework achieves 48.3% accuracy.<n>Results on SuperGPQA and TRQA confirm robustness across domains.
  arXiv  Detail & Related papers  (2025-09-25T14:05:55Z)
• AEGIS : Automated Co-Evolutionary Framework for Guarding Prompt Injections Schema [39.44407870355891]
  We propose AEGIS, an automated co-Evolutionary framework for Guarding prompt Injections.<n>Both attack and defense prompts are iteratively optimized against each other using a gradient-like natural language prompt optimization technique.<n>We evaluate our system on a real-world assignment grading dataset of prompt injection attacks and demonstrate that our method consistently outperforms existing baselines.
  arXiv  Detail & Related papers  (2025-08-27T12:25:45Z)
• Enhancing Robustness of LLM-Driven Multi-Agent Systems through Randomized Smoothing [13.997409139696556]
  This paper presents a framework for enhancing the safety of large language model (LLM) empowered multi-agent systems (MAS) in safety-critical domains such as aerospace.<n>We apply randomized smoothing, a statistical robustness certification technique, to the MAS consensus context, enabling probabilistic guarantees on agent decisions under adversarial influence.
  arXiv  Detail & Related papers  (2025-07-05T17:26:08Z)
• AegisLLM: Scaling Agentic Systems for Self-Reflective Defense in LLM Security [74.22452069013289]
  AegisLLM is a cooperative multi-agent defense against adversarial attacks and information leakage.<n>We show that scaling agentic reasoning system at test-time substantially enhances robustness without compromising model utility.<n> Comprehensive evaluations across key threat scenarios, including unlearning and jailbreaking, demonstrate the effectiveness of AegisLLM.
  arXiv  Detail & Related papers  (2025-04-29T17:36:05Z)
• Zero-Shot Defense Against Toxic Images via Inherent Multimodal Alignment in LVLMs [10.463762448166714]
  We introduce SafeCLIP, a lightweight method that leverages LVLMs inherent multimodal alignment for zero-shot toxic image detection.<n>Experiments show that SafeCLIP achieves a 66.9% defense success rate with only 3.2% false positive rate and 7.2% overhead.<n>Our work demonstrates that leveraging inherent multimodal alignment can yield efficient, low-cost LVLM safety.
  arXiv  Detail & Related papers  (2025-02-25T06:51:16Z)
• SafeAgentBench: A Benchmark for Safe Task Planning of Embodied LLM Agents [58.65256663334316]
  We present SafeAgentBench -- the first benchmark for safety-aware task planning of embodied LLM agents in interactive simulation environments.<n>SafeAgentBench includes: (1) an executable, diverse, and high-quality dataset of 750 tasks, rigorously curated to cover 10 potential hazards and 3 task types; (2) SafeAgentEnv, a universal embodied environment with a low-level controller, supporting multi-agent execution with 17 high-level actions for 9 state-of-the-art baselines; and (3) reliable evaluation methods from both execution and semantic perspectives.
  arXiv  Detail & Related papers  (2024-12-17T18:55:58Z)
• Palisade -- Prompt Injection Detection Framework [0.9620910657090188]
  Large Language Models are vulnerable to malicious prompt injection attacks. This paper proposes a novel NLP based approach for prompt injection detection. It emphasizes accuracy and optimization through a layered input screening process.
  arXiv  Detail & Related papers  (2024-10-28T15:47:03Z)
• AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases [73.04652687616286]
  We propose AgentPoison, the first backdoor attack targeting generic and RAG-based LLM agents by poisoning their long-term memory or RAG knowledge base. Unlike conventional backdoor attacks, AgentPoison requires no additional model training or fine-tuning. On each agent, AgentPoison achieves an average attack success rate higher than 80% with minimal impact on benign performance.
  arXiv  Detail & Related papers  (2024-07-17T17:59:47Z)
• ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models [65.79770974145983]
  ASSERT, Automated Safety Scenario Red Teaming, consists of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. We find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings.
  arXiv  Detail & Related papers  (2023-10-14T17:10:28Z)
• Distributed Dynamic Safe Screening Algorithms for Sparse Regularization [73.85961005970222]
  We propose a new distributed dynamic safe screening (DDSS) method for sparsity regularized models and apply it on shared-memory and distributed-memory architecture respectively. We prove that the proposed method achieves the linear convergence rate with lower overall complexity and can eliminate almost all the inactive features in a finite number of iterations almost surely.
  arXiv  Detail & Related papers  (2022-04-23T02:45:55Z)
• Multi-agent Policy Optimization with Approximatively Synchronous Advantage Estimation [55.96893934962757]
  In multi-agent system, polices of different agents need to be evaluated jointly. In current methods, value functions or advantage functions use counter-factual joint actions which are evaluated asynchronously. In this work, we propose the approximatively synchronous advantage estimation.
  arXiv  Detail & Related papers  (2020-12-07T07:29:19Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.