SIGMA introduces skill-incidence graphs to compose agents from reusable skills, yielding higher average performance and robustness than topology-only baselines on reasoning and coding benchmarks.
hub Canonical reference
G-designer: Architecting multi-agent communication topologies via graph neural networks
Canonical reference. 80% of citing Pith papers cite this work as background.
18
Pith papers citing it
1
external citations · Pith
Background
80% of classified citations
abstract
Recent advancements in large language model (LLM)-based agents have demonstrated that collective intelligence can significantly surpass the capabilities of individual agents, primarily due to well-crafted inter-agent communication topologies. Despite the diverse and high-performing designs available, practitioners often face confusion when selecting the most effective pipeline for their specific task: \textit{Which topology is the best choice for my task, avoiding unnecessary communication token overhead while ensuring high-quality solution?} In response to this dilemma, we introduce G-Designer, an adaptive, efficient, and robust solution for multi-agent deployment, which dynamically designs task-aware, customized communication topologies. Specifically, G-Designer models the multi-agent system as a multi-agent network, leveraging a variational graph auto-encoder to encode both the nodes (agents) and a task-specific virtual node, and decodes a task-adaptive and high-performing communication topology. Extensive experiments on six benchmarks showcase that G-Designer is: \textbf{(1) high-performing}, achieving superior results on MMLU with accuracy at $84.50\%$ and on HumanEval with pass@1 at $89.90\%$; \textbf{(2) task-adaptive}, architecting communication protocols tailored to task difficulty, reducing token consumption by up to $95.33\%$ on HumanEval; and \textbf{(3) adversarially robust}, defending against agent adversarial attacks with merely $0.3\%$ accuracy drop.
hub tools
citation-role summary
background 4
baseline 1
citation-polarity summary
representative citing papers
Geometry-aware FGW prior transfer plus PAC-Bayes residual adaptation reduces topology forgetting and raises average accuracy across continual multi-agent topology learning streams.
A new benchmark study finds that prompt optimization can deliver significant gains in multi-agent LLM systems but its effectiveness varies strongly with task, workflow, communication protocol, and team size.
MetaPS trains models via simulation rollouts to select from programmatic strategy libraries for market agents, yielding better performance than fixed or direct LLM baselines across model sizes.
TaskWeave enables LLM agents to sustain coherent long-horizon organizational dynamics in a simulated year-long IT company via structured memory and planning cycles.
PEAR is a permutation-equivariant adaptive routing protocol for multi-agent LLM debate that reconfigures sparse topologies each round to improve accuracy over fixed debate baselines.
LEMON trains an LLM orchestrator with counterfactual-augmented GRPO to produce deployable multi-agent specifications that reach state-of-the-art results on six reasoning and coding benchmarks.
EvoMAS trains a workflow adapter with policy gradients to dynamically instantiate stage-specific multi-agent workflows from a fixed agent pool, using explicit task-state construction and terminal success signals, and outperforms static baselines on GAIA, HLE, and DeepResearcher.
AgentCollabBench shows that multi-agent reliability is limited by communication topology, with converging-DAG nodes causing synthesis bottlenecks that discard constraints and explain 7-40% of information loss variance.
An ensemble-based information-theoretic active learning method using ensemble Kalman inversion selects valuable tasks to optimize communication structures in LLM multi-agent systems more reliably than random sampling under limited training budgets.
Translating historical governance into LLM multi-agent systems shows institutional topology drives collective performance gaps over 57 points, with optimal forms shifting by model capability and task.
SkillGraph jointly evolves agent skills and collaboration topologies in multi-agent vision-language systems using a multimodal graph transformer and a skill designer, yielding consistent performance gains on benchmarks.
Complete cyclic subtask graphs offer a lens to measure when multi-agent revisitation aids recovery and exploration versus when it increases costs or is dominated by other bottlenecks in LLM agent workflows.
SGH replaces implicit agent loops with explicit static DAGs, immutable execution plans, layered planning/recovery, and strict escalation protocols to improve controllability in LLM agents.
Evo-Memory is a new streaming benchmark and evaluation framework for self-evolving memory in LLM agents, unifying over ten memory modules and introducing the ReMem pipeline for continual improvement on multi-turn and reasoning datasets.
Structural diversity—how many disconnected prior collaboration communities a team bridges—predicts disruptive scientific impact better than team freshness or density and can offset the large-team incrementalism penalty.
Graphs can help LLMs reduce hallucinations, boost reasoning via prompting techniques, and better process structured data.
A survey unifying the operator-theoretic, probabilistic, and optimization-based mathematical structures underlying modern reinforcement learning algorithms.
citing papers explorer
-
SIGMA: Skill-Incidence Graphs for Compositional Multi-Agent Design
SIGMA introduces skill-incidence graphs to compose agents from reusable skills, yielding higher average performance and robustness than topology-only baselines on reasoning and coding benchmarks.
-
MasFACT: Continual Multi-Agent Topology Learning via Geometry-Aware Posterior Transfer
Geometry-aware FGW prior transfer plus PAC-Bayes residual adaptation reduces topology forgetting and raises average accuracy across continual multi-agent topology learning streams.
-
MAS-PromptBench: When Does Prompt Optimization Improve Multi-Agent LLM Systems?
A new benchmark study finds that prompt optimization can deliver significant gains in multi-agent LLM systems but its effectiveness varies strongly with task, workflow, communication protocol, and team size.
-
MetaPS: Adaptive Programmatic Strategy Selection for Market Agents
MetaPS trains models via simulation rollouts to select from programmatic strategy libraries for market agents, yielding better performance than fixed or direct LLM baselines across model sizes.
-
Can LLM Agents Sustain Long-Horizon Organizational Dynamics?
TaskWeave enables LLM agents to sustain coherent long-horizon organizational dynamics in a simulated year-long IT company via structured memory and planning cycles.
-
PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate
PEAR is a permutation-equivariant adaptive routing protocol for multi-agent LLM debate that reconfigures sparse topologies each round to improve accuracy over fixed debate baselines.
-
LEMON: Learning Executable Multi-Agent Orchestration via Counterfactual Reinforcement Learning
LEMON trains an LLM orchestrator with counterfactual-augmented GRPO to produce deployable multi-agent specifications that reach state-of-the-art results on six reasoning and coding benchmarks.
-
EvoMAS: Learning Execution-Time Workflows for Multi-Agent Systems
EvoMAS trains a workflow adapter with policy gradients to dynamically instantiate stage-specific multi-agent workflows from a fixed agent pool, using explicit task-state construction and terminal success signals, and outperforms static baselines on GAIA, HLE, and DeepResearcher.
-
AgentCollabBench: Diagnosing When Good Agents Make Bad Collaborators
AgentCollabBench shows that multi-agent reliability is limited by communication topology, with converging-DAG nodes causing synthesis bottlenecks that discard constraints and explain 7-40% of information loss variance.
-
Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent Systems
An ensemble-based information-theoretic active learning method using ensemble Kalman inversion selects valuable tasks to optimize communication structures in LLM multi-agent systems more reliably than random sampling under limited training budgets.
-
When Agents Evolve, Institutions Follow
Translating historical governance into LLM multi-agent systems shows institutional topology drives collective performance gaps over 57 points, with optimal forms shifting by model capability and task.
-
SkillGraph: Self-Evolving Multi-Agent Collaboration with Multimodal Graph Topology
SkillGraph jointly evolves agent skills and collaboration topologies in multi-agent vision-language systems using a multimodal graph transformer and a skill designer, yielding consistent performance gains on benchmarks.
-
Complete Cyclic Subtask Graphs for Tool-Using LLM Agents: Flexibility, Cost, and Bottlenecks in Multi-Agent Workflows
Complete cyclic subtask graphs offer a lens to measure when multi-agent revisitation aids recovery and exploration versus when it increases costs or is dominated by other bottlenecks in LLM agent workflows.
-
From Agent Loops to Structured Graphs:A Scheduler-Theoretic Framework for LLM Agent Execution
SGH replaces implicit agent loops with explicit static DAGs, immutable execution plans, layered planning/recovery, and strict escalation protocols to improve controllability in LLM agents.
-
Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory
Evo-Memory is a new streaming benchmark and evaluation framework for self-evolving memory in LLM agents, unifying over ten memory modules and introducing the ReMem pipeline for continual improvement on multi-turn and reasoning datasets.
-
SP-GCRL: Influence Maximization on Incomplete Social Graphs
Structural diversity—how many disconnected prior collaboration communities a team bridges—predicts disruptive scientific impact better than team freshness or density and can offset the large-team incrementalism penalty.
-
Position: How can Graphs Help Large Language Models?
Graphs can help LLMs reduce hallucinations, boost reasoning via prompting techniques, and better process structured data.
-
Mathematical methods of reinforcement learning
A survey unifying the operator-theoretic, probabilistic, and optimization-based mathematical structures underlying modern reinforcement learning algorithms.