Imagen de Portada

SABER: Small Actions, Big Errors -- Safeguarding Mutating Steps in LLM Agents

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Cuadron, Alejandro, Yu, Pengfei, Liu, Yang, Gupta, Arpit
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Hierarchical Reinforcement Learning with Augmented Step-Level Transitions for LLM Agents
por: Zhen, Shuai, et al.
Publicado: (2026)

HashAttention: Semantic Sparsity for Faster Inference
por: Desai, Aditya, et al.
Publicado: (2024)

Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment
por: Wang, Haozhong, et al.
Publicado: (2026)

CORA: Conformal Risk-Controlled Agents for Safeguarded Mobile GUI Automation
por: Feng, Yushi, et al.
Publicado: (2026)

Learnware of Language Models: Specialized Small Language Models Can Do Big
por: Tan, Zhi-Hao, et al.
Publicado: (2025)

RAMP: Reinforcement Adaptive Mixed Precision Quantization for Efficient On Device LLM Inference
por: Gautam, Arpit Singh, et al.
Publicado: (2026)

GuardReasoner: Towards Reasoning-based LLM Safeguards
por: Liu, Yue, et al.
Publicado: (2025)

JudgeBench: A Benchmark for Evaluating LLM-based Judges
por: Tan, Sijun, et al.
Publicado: (2024)

vAttention: Verified Sparse Attention
por: Desai, Aditya, et al.
Publicado: (2025)

SABER: Switchable and Balanced Training for Efficient LLM Reasoning
por: Zhao, Kai, et al.
Publicado: (2025)

Winning Big with Small Models: Knowledge Distillation vs. Self-Training for Reducing Hallucination in Product QA Agents
por: Lewis, Ashley, et al.
Publicado: (2025)

Verifiability-First Agents: Provable Observability and Lightweight Audit Agents for Controlling Autonomous LLM Systems
por: Gupta, Abhivansh
Publicado: (2025)

Bootstrapped Mixed Rewards for RL Post-Training: Injecting Canonical Action Order
por: Gupta, Prakhar, et al.
Publicado: (2025)

Watch Every Step! LLM Agent Learning via Iterative Step-Level Process Refinement
por: Xiong, Weimin, et al.
Publicado: (2024)

One-Step Graph-Structured Neural Flows for Irregular Multivariate Time Series Classification
por: Gao, Mengzhou, et al.
Publicado: (2026)

When LLM Meets Time Series: Can LLMs Perform Multi-Step Time Series Reasoning and Inference
por: Ye, Wen, et al.
Publicado: (2025)

QDeepGR4J: Quantile-based ensemble of deep learning and GR4J hybrid rainfall-runoff models for extreme flow prediction with uncertainty quantification
por: Kapoor, Arpit, et al.
Publicado: (2025)

Uncertainty in Action: Confidence Elicitation in Embodied Agents
por: Yu, Tianjiao, et al.
Publicado: (2025)

Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents
por: Song, Yifan, et al.
Publicado: (2024)

SPARK: Stepwise Process-Aware Rewards for Reference-Free Reinforcement Learning
por: Rahman, Salman, et al.
Publicado: (2025)

An Example Safety Case for Safeguards Against Misuse
por: Clymer, Joshua, et al.
Publicado: (2025)

STeCa: Step-level Trajectory Calibration for LLM Agent Learning
por: Wang, Hanlin, et al.
Publicado: (2025)

Two-Step Offline Preference-Based Reinforcement Learning with Constrained Actions
por: Xu, Yinglun, et al.
Publicado: (2023)

GoAgent: Group-of-Agents Communication Topology Generation for LLM-based Multi-Agent Systems
por: Chen, Hongjiang, et al.
Publicado: (2026)

PoPE: Legendre Orthogonal Polynomials Based Position Encoding for Large Language Models
por: Aggarwal, Arpit
Publicado: (2024)

PropGuard: Safeguarding LLM-MAS via Propagation-Aware Exploration and Remediation
por: Yan, Bingyu, et al.
Publicado: (2026)

Bellman Error Centering
por: Chen, Xingguo, et al.
Publicado: (2025)

PFGuard: A Generative Framework with Privacy and Fairness Safeguards
por: Kim, Soyeon, et al.
Publicado: (2024)

Multi-Agent Decision Transformers for Dynamic Dispatching in Material Handling Systems Leveraging Enterprise Big Data
por: Lee, Xian Yeow, et al.
Publicado: (2024)

Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation
por: Halawi, Danny, et al.
Publicado: (2024)

LLM Reasoning with Process Rewards for Outcome-Guided Steps
por: Rezaei, Mohammad, et al.
Publicado: (2026)

Accordion-Thinking: Self-Regulated Step Summaries for Efficient and Readable LLM Reasoning
por: Yang, Zhicheng, et al.
Publicado: (2026)

FT-Dojo: Towards Autonomous LLM Fine-Tuning with Language Agents
por: Li, Qizheng, et al.
Publicado: (2026)

Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation
por: Ding, Ziyi, et al.
Publicado: (2026)

Agent-Omit: Adaptive Context Omission for Efficient LLM Agents
por: Ning, Yansong, et al.
Publicado: (2026)

GROW: Aligning GRPO with State-Action Modeling for Open-World VLM Agents
por: Wu, Xiongbin, et al.
Publicado: (2026)

Think like a Scientist: Physics-guided LLM Agent for Equation Discovery
por: Yang, Jianke, et al.
Publicado: (2026)

On Evaluating LLM Alignment by Evaluating LLMs as Judges
por: Liu, Yixin, et al.
Publicado: (2025)

Agentic Unlearning: When LLM Agent Meets Machine Unlearning
por: Wang, Bin, et al.
Publicado: (2026)

Reinforcement Learning With Sparse-Executing Actions via Sparsity Regularization
por: Pang, Jing-Cheng, et al.
Publicado: (2021)