SimWorld Studio deploys an evolving coding agent to create adaptive 3D environments that co-evolve with embodied learners, delivering 18-point success-rate gains over fixed environments in navigation benchmarks.
hub Mixed citations
Osworld: Benchmarking multimodal agents for open-ended tasks in real computer environments.Advances in Neural Information Processing Systems, 37:52040–52094
Mixed citation behavior. Most common role is background (62%).
25
Pith papers citing it
Background
62% of classified citations
hub tools
citation-role summary
background 5
dataset 3
citation-polarity summary
representative citing papers
A²utoLPBench is a generator that produces unlimited LP word problems with ground-truth answers known by construction via inverse-KKT, bundled with a Docker environment for agent evaluation.
Proposes COM-as-Action paradigm for deterministic software manipulation, introduces ComCADBench benchmark and ComActor agent that achieves SOTA performance over GUI baselines.
MemGym unifies agent gyms into a memory benchmark with isolated scoring across tool-use, research, coding, and computer-use regimes plus a lightweight reward model for tractable coding evaluation.
Introduces the 1GC-7RC benchmark to evaluate AI coding agents on seven diverse ML tasks under single-GPU time and access constraints.
Coding agents struggle to infer least-privilege file permissions by omitting needed accesses while granting unused or sensitive ones, but Sufficiency-Tightness Decomposition improves sensitive-task success by up to 15.8% and reduces attacks.
FIKA-Bench is a leakage-aware benchmark of 311 instances showing that even the best large multimodal models and tool-equipped agents reach only 25.1% accuracy on fine-grained recognition questions that require external evidence search and verification.
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
A VOI-based controller for dual inference budgets improves multi-hop QA performance by prioritizing search actions and selectively finalizing answers.
ClawMark is a new benchmark for multi-turn multi-day multimodal coworker agents in stateful evolving services, with deterministic Python checkers showing frontier models achieve only 20% strict task success.
A reward model trained on 1,130 human preference judgments outperforms base agents by 120 Elo points on a 50-task benchmark for recovering from harmful states in computer-use environments.
RiskWebWorld is the first realistic interactive benchmark for GUI agents in e-commerce risk management, revealing a large gap between generalist and specialized models plus RL gains.
A capability-driven benchmark of 400 bilingual real-world tasks shows current proactive agents fail >50% of the time, with framework architecture impacting performance more than base model choice.
COMFYCLAW introduces skill evolution via graph editing, automatic reversion, VLM verification, and distillation of runs into reusable Agent Skills, achieving higher average scores than a verifier-only baseline across benchmarks.
SEAL co-evolves LLM agents and environments via shared turn-level failure diagnoses, yielding +8.25 to +26.25 point gains on tool-use tasks with only 400 samples.
SynAE is a multi-metric framework that evaluates how well synthetic benchmarks replicate real data characteristics for multi-turn tool-calling agent testing.
AQuaUI uses adaptive quadtrees to cut visual tokens in GUI-agent LMMs by up to 29.52% at inference time while retaining 99.06% of full-token accuracy on grounding and navigation benchmarks.
MementoGUI introduces a modular memory-control framework with working and episodic memory operators that improves long-horizon GUI agent performance over history-replay and text-only baselines.
ToolCUA introduces a trajectory scaling pipeline and staged RL to optimize GUI-tool switching, reaching 46.85% accuracy on OSWorld-MCP for a 66% relative gain over baseline.
FATE lets LLM agents self-evolve safer behaviors by generating and filtering repairs from their own failure trajectories using verifiers and Pareto optimization.
DTap is a new red-teaming platform for AI agents that uses autonomous exploration across realistic simulations to discover vulnerabilities and creates a verifiable benchmark dataset.
Personal agents require edge deployment to preserve high-fidelity local context and zero-latency loops, as claimed through three structural shifts away from cloud-centric designs.
The paper develops a unified framework that organizes computer-use agent reliability around perception-decision-execution layers and creation-deployment-operation-maintenance stages to map security and alignment interventions.
A survey that taxonomizes threats to agentic AI, reviews benchmarks and evaluation methods, discusses technical and governance defenses, and identifies open challenges.
citing papers explorer
-
SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning
SimWorld Studio deploys an evolving coding agent to create adaptive 3D environments that co-evolve with embodied learners, delivering 18-point success-rate gains over fixed environments in navigation benchmarks.
-
A$^{2}$utoLPBench: An Auto-Generated, Agent-Friendly LP Benchmark via Inverse-KKT Construction
A²utoLPBench is a generator that produces unlimited LP word problems with ground-truth answers known by construction via inverse-KKT, bundled with a Docker environment for agent evaluation.
-
ComAct: Reframing Professional Software Manipulation via COM-as-Action Paradigm
Proposes COM-as-Action paradigm for deterministic software manipulation, introduces ComCADBench benchmark and ComActor agent that achieves SOTA performance over GUI baselines.
-
MemGym: a Long-Horizon Memory Environment for LLM Agents
MemGym unifies agent gyms into a memory benchmark with isolated scoring across tool-use, research, coding, and computer-use regimes plus a lightweight reward model for tractable coding evaluation.
-
1GC-7RC: One Graphic Card -- Seven Research Challenges! How Good Are AI Agents at Doing Your Job?
Introduces the 1GC-7RC benchmark to evaluate AI coding agents on seven diverse ML tasks under single-GPU time and access constraints.
-
Do Coding Agents Understand Least-Privilege Authorization?
Coding agents struggle to infer least-privilege file permissions by omitting needed accesses while granting unused or sensitive ones, but Sufficiency-Tightness Decomposition improves sensitive-task success by up to 15.8% and reduces attacks.
-
FIKA-Bench: From Fine-grained Recognition to Fine-Grained Knowledge Acquisition
FIKA-Bench is a leakage-aware benchmark of 311 instances showing that even the best large multimodal models and tool-equipped agents reach only 25.1% accuracy on fine-grained recognition questions that require external evidence search and verification.
-
Covering Human Action Space for Computer Use: Data Synthesis and Benchmark
Presents CUActSpot benchmark and renderer-LLM data synthesis that lets a 4B model outperform larger open-source models on complex computer interactions.
-
Inference-Time Budget Control for LLM Search Agents
A VOI-based controller for dual inference budgets improves multi-hop QA performance by prioritizing search actions and selectively finalizing answers.
-
ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents
ClawMark is a new benchmark for multi-turn multi-day multimodal coworker agents in stateful evolving services, with deterministic Python checkers showing frontier models achieve only 20% strict task success.
-
Human-Guided Harm Recovery for Computer Use Agents
A reward model trained on 1,130 human preference judgments outperforms base agents by 120 Elo points on a 50-task benchmark for recovering from harmful states in computer-use environments.
-
RiskWebWorld: A Realistic Interactive Benchmark for GUI Agents in E-commerce Risk Management
RiskWebWorld is the first realistic interactive benchmark for GUI agents in e-commerce risk management, revealing a large gap between generalist and specialized models plus RL gains.
-
UniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks
A capability-driven benchmark of 400 bilingual real-world tasks shows current proactive agents fail >50% of the time, with framework architecture impacting performance more than base model choice.
-
COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows
COMFYCLAW introduces skill evolution via graph editing, automatic reversion, VLM verification, and distillation of runs into reusable Agent Skills, achieving higher average scores than a verifier-only baseline across benchmarks.
-
SEAL: Synergistic Co-Evolution of Agents and Learning Environments
SEAL co-evolves LLM agents and environments via shared turn-level failure diagnoses, yielding +8.25 to +26.25 point gains on tool-use tasks with only 400 samples.
-
SynAE: A Framework for Measuring the Quality of Synthetic Data for Tool-Calling Agent Evaluations
SynAE is a multi-metric framework that evaluates how well synthetic benchmarks replicate real data characteristics for multi-turn tool-calling agent testing.
-
AQuaUI: Visual Token Reduction for GUI Agents with Adaptive Quadtrees
AQuaUI uses adaptive quadtrees to cut visual tokens in GUI-agent LMMs by up to 29.52% at inference time while retaining 99.06% of full-token accuracy on grounding and navigation benchmarks.
-
MementoGUI: Learning Agentic Multimodal Memory Control for Long-Horizon GUI Agents
MementoGUI introduces a modular memory-control framework with working and episodic memory operators that improves long-horizon GUI agent performance over history-replay and text-only baselines.
-
ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agents
ToolCUA introduces a trajectory scaling pipeline and staged RL to optimize GUI-tool switching, reaching 46.85% accuracy on OSWorld-MCP for a 66% relative gain over baseline.
-
On-Policy Self-Evolution via Failure Trajectories for Agentic Safety Alignment
FATE lets LLM agents self-evolve safer behaviors by generating and filtering repairs from their own failure trajectories using verifiers and Pareto optimization.
-
DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents
DTap is a new red-teaming platform for AI agents that uses autonomous exploration across realistic simulations to discover vulnerabilities and creates a verifiable benchmark dataset.
-
Beyond Scaling: Agents Are Heading to the Edge
Personal agents require edge deployment to preserve high-fidelity local context and zero-latency loops, as claimed through three structural shifts away from cloud-centric designs.
-
Securing Computer-Use Agents: A Unified Architecture-Lifecycle Framework for Deployment-Grounded Reliability
The paper develops a unified framework that organizes computer-use agent reliability around perception-decision-execution layers and creation-deployment-operation-maintenance stages to map security and alignment interventions.
-
Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges
A survey that taxonomizes threats to agentic AI, reviews benchmarks and evaluation methods, discusses technical and governance defenses, and identifies open challenges.
-
Skills-Coach: A Self-Evolving Skill Optimizer via Training-Free GRPO
Skills-Coach optimizes LLM agent skills via task generation, prompt/code tuning, comparative execution, and traceable evaluation, reporting gains on a 48-skill benchmark called Skill-X.