TerraBench is a new benchmark with 403 tasks across Earth-science domains that evaluates LLM agents on coordinating heterogeneous data using executable ReAct-style workflows and process-level metrics.
hub
AgentScope: A Flexible yet Robust Multi-Agent Platform
22 Pith papers cite this work, alongside 9 external citations. Polarity classification is still indexing.
22
Pith papers citing it
9
external citations · external index
hub tools
citation-role summary
background 2
method 1
citation-polarity summary
representative citing papers
S-Bus reconstructs read sets from HTTP traffic for multi-agent LLM state coordination, delivering Observable-Read Isolation with formal proofs and empirical safety matching traditional databases.
HAM³ achieves up to 78.3% attack success rate on the GQA benchmark by hierarchically attacking perception, communication, and reasoning layers in multi-modal multi-agent systems.
DESBench reveals structural trade-offs among centralized, hierarchical, heterarchical, and holonic coordination in dynamic industrial scheduling that outcome metrics alone miss.
TS-Reasoner is a domain-oriented agent using LLMs, computational tools, and error feedback for multi-step time series inference, showing better performance than general LLMs on understanding and reasoning benchmarks.
OrchestrXR uses multi-agent orchestration with structured schemas to generate Unity XR study prototypes from ideas, supported by a user study with 12 researchers indicating effective support and intent preservation.
GenWorld supplies a data-grounded synthetic urban environment, structured agent interface, and offline LLM policy compilation to enable scalable city-scale LLM-agent simulations, shown via three cases in Higashihiroshima with census and mobile-data checks.
A client-server swarm framework lets arbitrary agent code run on detached machines while shared servers train it, and context-timeline merging cuts actor-update time by 6.25x on AppWorld.
PatchBoard introduces schema-grounded JSON Patch state mutations with an Architect agent and validation kernel, reporting 84.6% success and lower token use on 630 ALFWorld episodes versus LangGraph and Flock baselines.
Agent Capsules is an adaptive runtime for multi-agent LLM pipelines that selectively compounds agent executions under quality constraints, delivering 19-68% token reductions at parity or better quality versus LangGraph and DSPy baselines.
Multi-agent systems amplify minor stochastic biases into systemic polarization via echo-chamber effects in structured workflows, even with neutral agents.
TokenCake introduces agent-aware temporal and spatial schedulers for KV cache management in LLM multi-agent serving, claiming over 47% lower end-to-end latency and up to 16.9% better GPU memory utilization than vLLM on representative benchmarks.
BlindGuard introduces an unsupervised hierarchical agent encoder plus corruption-guided contrastive detector that identifies malicious agents in LLM-based multi-agent systems without any attack labels or prior knowledge of malicious behaviors.
Introduces six-dimension trustworthiness definition and attention-based A-Trust score with a TMS to improve LLM-MAS robustness against malicious or unreliable messages.
AgentSociety is a large-scale LLM agent-based social simulator validated on polarization, UBI, disasters, and sustainability issues with alignment to real experiments.
Large-scale empirical study finds widespread prompt leaking in commercial LLM apps and introduces AREA defense that improves usability while resisting leaks.
RAC is a log-based recovery paradigm implemented as an architectural extension to agent frameworks, achieving 1.5-8X better latency and token economy than LLM-based recovery on τ-bench and REALM-Bench.
Proposes and tests a constitutive definition of 'agent harness' via conceptual analysis of literature and six real systems.
SDOF combines an RLHF-trained intent router with a state-aware dispatcher using finite automata to constrain multi-agent orchestration, reporting 80.9% routing accuracy and 86.5% task completion on a recruitment platform while blocking unsafe actions.
InfantAgent-Next integrates tool-based and vision agents in a modular architecture and reports 7.27% accuracy on OSWorld, exceeding Claude-Computer-Use while also testing on GAIA and SWE-Bench.
A literature survey that collects and categorizes 124 papers on LLM-based agents for software engineering from SE and agent perspectives.
citing papers explorer
-
TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?
TerraBench is a new benchmark with 403 tasks across Earth-science domains that evaluates LLM agents on coordinating heterogeneous data using executable ReAct-style workflows and process-level metrics.
-
S-Bus: Automatic Read-Set Reconstruction for Multi-Agent LLM State Coordination
S-Bus reconstructs read sets from HTTP traffic for multi-agent LLM state coordination, delivering Observable-Read Isolation with formal proofs and empirical safety matching traditional databases.
-
Hierarchical Attacks for Multi-Modal Multi-Agent Reasoning
HAM³ achieves up to 78.3% attack success rate on the GQA benchmark by hierarchically attacking perception, communication, and reasoning layers in multi-modal multi-agent systems.
-
When Does Hierarchy Help? Benchmarking Agent Coordination in Event-Driven Industrial Scheduling
DESBench reveals structural trade-offs among centralized, hierarchical, heterarchical, and holonic coordination in dynamic industrial scheduling that outcome metrics alone miss.
-
TS-Reasoner: Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis
TS-Reasoner is a domain-oriented agent using LLMs, computational tools, and error feedback for multi-step time series inference, showing better performance than general LLMs on understanding and reasoning benchmarks.
-
OrchestrXR: A Multi-Agent System for Idea-to-Prototype XR Study Authoring
OrchestrXR uses multi-agent orchestration with structured schemas to generate Unity XR study prototypes from ideas, supported by a user study with 12 researchers indicating effective support and intent preservation.
-
GenWorld: Empirically Grounded Urban Simulation Infrastructure for Scalable LLM-Agent Studies
GenWorld supplies a data-grounded synthetic urban environment, structured agent interface, and offline LLM policy compilation to enable scalable city-scale LLM-agent simulations, shown via three cases in Higashihiroshima with census and mobile-data checks.
-
AgentJet: A Distributed Swarm Training Framework for Agentic Reinforcement Learning
A client-server swarm framework lets arbitrary agent code run on detached machines while shared servers train it, and context-timeline merging cuts actor-update time by 6.25x on AppWorld.
-
PatchBoard: Schema-Grounded State Mutation for Reliable and Auditable LLM Multi-Agent Collaboration
PatchBoard introduces schema-grounded JSON Patch state mutations with an Architect agent and validation kernel, reporting 84.6% success and lower token use on 630 ALFWorld episodes versus LangGraph and Flock baselines.
-
Agent Capsules: Quality-Gated Granularity Control for Multi-Agent LLM Pipelines
Agent Capsules is an adaptive runtime for multi-agent LLM pipelines that selectively compounds agent executions under quality constraints, delivering 19-68% token reductions at parity or better quality versus LangGraph and DSPy baselines.
-
Aligned Agents, Biased Swarm: Measuring Bias Amplification in Multi-Agent Systems
Multi-agent systems amplify minor stochastic biases into systemic polarization via echo-chamber effects in structured workflows, even with neutral agents.
-
TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications
TokenCake introduces agent-aware temporal and spatial schedulers for KV cache management in LLM multi-agent serving, claiming over 47% lower end-to-end latency and up to 16.9% better GPU memory utilization than vLLM on representative benchmarks.
-
BlindGuard: Safeguarding LLM-based Multi-Agent Systems under Unknown Attacks
BlindGuard introduces an unsupervised hierarchical agent encoder plus corruption-guided contrastive detector that identifies malicious agents in LLM-based multi-agent systems without any attack labels or prior knowledge of malicious behaviors.
-
To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems
Introduces six-dimension trustworthiness definition and attention-based A-Trust score with a TMS to improve LLM-MAS robustness against malicious or unreliable messages.
-
AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society
AgentSociety is a large-scale LLM agent-based social simulator validated on polarization, UBI, disasters, and sustainability issues with alignment to real experiments.
-
Understanding and Mitigating Prompt Leaking Attacks in Real-World LLM-Based Applications
Large-scale empirical study finds widespread prompt leaking in commercial LLM apps and introduces AREA defense that improves usability while resisting leaks.
-
Robust Agent Compensation (RAC): Teaching AI Agents to Compensate
RAC is a log-based recovery paradigm implemented as an architectural extension to agent frameworks, achieving 1.5-8X better latency and token economy than LLM-based recovery on τ-bench and REALM-Bench.
-
What makes a harness a harness: necessary and sufficient conditions for an agent harness
Proposes and tests a constitutive definition of 'agent harness' via conceptual analysis of literature and six real systems.
-
SDOF: Taming the Alignment Tax in Multi-Agent Orchestration with State-Constrained Dispatch
SDOF combines an RLHF-trained intent router with a state-aware dispatcher using finite automata to constrain multi-agent orchestration, reporting 80.9% routing accuracy and 86.5% task completion on a recruitment platform while blocking unsafe actions.
-
InfantAgent-Next: A Multimodal Generalist Agent for Automated Computer Interaction
InfantAgent-Next integrates tool-based and vision agents in a modular architecture and reports 7.27% accuracy on OSWorld, exceeding Claude-Computer-Use while also testing on GAIA and SWE-Bench.
-
Large Language Model-Based Agents for Software Engineering: A Survey
A literature survey that collects and categorizes 124 papers on LLM-based agents for software engineering from SE and agent perspectives.
- Agent libOS: A Runtime Substrate for Capability-Controlled Self-Evolving LLM Agents