Imagen de Portada

TRACE: Trajectory Risk-Aware Compression for Long-Horizon Agent Safety

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Hong, Zhepei, Wang, Lin, Li, Liting, Ma, Haokai, Fang, Junfeng, Shen, Fei, Zhang, Dan, Wang, Xiang
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems
por: Fang, Junfeng, et al.
Publicado: (2025)

Beyond Reactive Safety: Risk-Aware LLM Alignment via Long-Horizon Simulation
por: Sun, Chenkai, et al.
Publicado: (2025)

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios
por: Luo, Haotian, et al.
Publicado: (2025)

LPS-Bench: Benchmarking Safety Awareness of Computer-Use Agents in Long-Horizon Planning under Benign and Adversarial Scenarios
por: Chen, Tianyu, et al.
Publicado: (2026)

Rubric-based On-policy Distillation
por: Fang, Junfeng, et al.
Publicado: (2026)

TRACE: A Multi-Agent System for Autonomous Physical Reasoning for Seismology
por: Liu, Feng, et al.
Publicado: (2026)

TRACE: Trajectory Recovery with State Propagation Diffusion for Urban Mobility
por: Wang, Jinming, et al.
Publicado: (2026)

ContextBudget: Budget-Aware Context Management for Long-Horizon Search Agents
por: Wu, Yong, et al.
Publicado: (2026)

Beyond Outcome Reward: Decoupling Search and Answering Improves LLM Agents
por: Wang, Yiding, et al.
Publicado: (2025)

Risk-Aware Vehicle Trajectory Prediction Under Safety-Critical Scenarios
por: Wang, Qingfan, et al.
Publicado: (2024)

AgentLAB: Benchmarking LLM Agents against Long-Horizon Attacks
por: Jiang, Tanqiu, et al.
Publicado: (2026)

The Cognitive Bandwidth Bottleneck: Shifting Long-Horizon Agent from Planning with Actions to Planning with Schemas
por: Xu, Baixuan, et al.
Publicado: (2025)

MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
por: Shi, Yaorui, et al.
Publicado: (2026)

R-Judge: Benchmarking Safety Risk Awareness for LLM Agents
por: Yuan, Tongxin, et al.
Publicado: (2024)

HiMem: Hierarchical Long-Term Memory for LLM Long-Horizon Agents
por: Zhang, Ningning, et al.
Publicado: (2026)

MemPO: Self-Memory Policy Optimization for Long-Horizon Agents
por: Li, Ruoran, et al.
Publicado: (2026)

RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation
por: Wang, Zihao, et al.
Publicado: (2024)

PII-Bench: Evaluating Query-Aware Privacy Protection Systems
por: Shen, Hao, et al.
Publicado: (2025)

From Cues to Horizons: Dynamic Risk Horizon Profiling for Trajectory Prediction
por: Ning, Xinyi, et al.
Publicado: (2026)

Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents
por: Chen, Zhuofu, et al.
Publicado: (2026)

Milestone-Guided Policy Learning for Long-Horizon Language Agents
por: Wang, Zixuan, et al.
Publicado: (2026)

LongSeeker: Elastic Context Orchestration for Long-Horizon Search Agents
por: Lu, Yijun, et al.
Publicado: (2026)

NExT-Guard: Training-Free Streaming Safeguard without Token-Level Labels
por: Fang, Junfeng, et al.
Publicado: (2026)

LeAdQA: LLM-Driven Context-Aware Temporal Grounding for Video Question Answering
por: Dong, Xinxin, et al.
Publicado: (2025)

Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents
por: Li, Yunfan, et al.
Publicado: (2026)

Efficient Long-Horizon GUI Agents via Training-Free KV Cache Compression
por: Zhou, Bowen, et al.
Publicado: (2026)

AgentFold: Long-Horizon Web Agents with Proactive Context Management
por: Ye, Rui, et al.
Publicado: (2025)

Hierarchical Two-Stage Framework for Environment-Aware Long-Horizon Vessel Trajectory Prediction
por: Gnanavel, Ganeshaaraj, et al.
Publicado: (2026)

ICA: Information-Aware Credit Assignment for Visually Grounded Long-Horizon Information-Seeking Agents
por: Pang, Cong, et al.
Publicado: (2026)

AgentCPM-Explore: Realizing Long-Horizon Deep Exploration for Edge-Scale Agents
por: Chen, Haotian, et al.
Publicado: (2026)

ACC: Compiling Agent Trajectories for Long-Context Training
por: Su, Qisheng, et al.
Publicado: (2026)

TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios
por: Shen, Yuanzhe, et al.
Publicado: (2026)

REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
por: Chu, Zheng, et al.
Publicado: (2026)

SAM: State-Adaptive Memory for Long-Horizon Reasoning Agent
por: Hu, Yuyang, et al.
Publicado: (2026)

QuarkMedSearch: A Long-Horizon Deep Search Agent for Exploring Medical Intelligence
por: Lin, Zhichao, et al.
Publicado: (2026)

HorizonBench: Long-Horizon Personalization with Evolving Preferences
por: Li, Shuyue Stella, et al.
Publicado: (2026)

SaaSBench: Exploring the Boundaries of Coding Agents in Long-Horizon Enterprise SaaS Engineering
por: Ren, Qingnan, et al.
Publicado: (2026)

ContextFlow: Hierarchical Task-State Alignment for Long-Horizon Embodied Agents
por: Guo, Shuhan, et al.
Publicado: (2026)

Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agents
por: Zhang, Yuxin, et al.
Publicado: (2026)

ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks
por: Song, Yuanyi, et al.
Publicado: (2025)