Imagen de Portada

Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Lu, Qingyu, Ding, Liang, Cao, Siyi, Liu, Xuebo, Zhang, Kanjian, Zhang, Jinxia, Tao, Dacheng
Formato:	Preprint
Publicado:	2025
Materias:	Computation and Language Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

The Bitter Lesson of Diffusion Language Models for Agentic Workflows: A Comprehensive Reality Check
por: Lu, Qingyu, et al.
Publicado: (2026)

MQM-APE: Toward High-Quality Error Annotation Predictors with Automatic Post-Editing in LLM Translation Evaluators
por: Lu, Qingyu, et al.
Publicado: (2024)

Error Analysis Prompting Enables Human-Like Translation Evaluation in Large Language Models
por: Lu, Qingyu, et al.
Publicado: (2023)

Self-Powered LLM Modality Expansion for Large Speech-Text Models
por: Yu, Tengfei, et al.
Publicado: (2024)

Exploring and Enhancing the Transfer of Distribution in Knowledge Distillation for Autoregressive Language Models
por: Rao, Jun, et al.
Publicado: (2024)

Revisiting Demonstration Selection Strategies in In-Context Learning
por: Peng, Keqin, et al.
Publicado: (2024)

Revisiting Catastrophic Forgetting in Large Language Model Tuning
por: Li, Hongyu, et al.
Publicado: (2024)

NEAT: Neuron-Based Early Exit for Large Reasoning Models
por: Liu, Kang, et al.
Publicado: (2026)

Intention Analysis Makes LLMs A Good Jailbreak Defender
por: Zhang, Yuqi, et al.
Publicado: (2024)

ROSE Doesn't Do That: Boosting the Safety of Instruction-Tuned Large Language Models with Reverse Prompt Contrastive Decoding
por: Zhong, Qihuang, et al.
Publicado: (2024)

EE-Tuning: An Economical yet Scalable Solution for Tuning Early-Exit Large Language Models
por: Pan, Xuchen, et al.
Publicado: (2024)

Accelerating Large Language Model Inference with Self-Supervised Early Exits
por: Valade, Florian
Publicado: (2024)

AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
por: Xi, Zhiheng, et al.
Publicado: (2024)

OOP: Object-Oriented Programming Evaluation Benchmark for Large Language Models
por: Wang, Shuai, et al.
Publicado: (2024)

AgentDropout: Dynamic Agent Elimination for Token-Efficient and High-Performance LLM-Based Multi-Agent Collaboration
por: Wang, Zhexuan, et al.
Publicado: (2025)

Early Exit Is a Natural Capability in Transformer-based Models: An Empirical Study on Early Exit without Joint Optimization
por: Shan, Weiqiao, et al.
Publicado: (2024)

Entropy-Guided Watermarking for LLMs: A Test-Time Framework for Robust and Traceable Text Generation
por: Cai, Shizhan, et al.
Publicado: (2025)

Uncertainty Aware Learning for Language Model Alignment
por: Wang, Yikun, et al.
Publicado: (2024)

AgentDropoutV2: Optimizing Information Flow in Multi-Agent Systems via Test-Time Rectify-or-Reject Pruning
por: Wang, Yutong, et al.
Publicado: (2026)

NoVo: Norm Voting off Hallucinations with Attention Heads in Large Language Models
por: Ho, Zheng Yi, et al.
Publicado: (2024)

Defending against Jailbreak through Early Exit Generation of Large Language Models
por: Zhao, Chongwen, et al.
Publicado: (2024)

FlashThink: An Early Exit Method For Efficient Reasoning
por: Jiang, Guochao, et al.
Publicado: (2025)

ADAM: An Embodied Causal Agent in Open-World Environments
por: Yu, Shu, et al.
Publicado: (2024)

E2S2: Encoding-Enhanced Sequence-to-Sequence Pretraining for Language Understanding and Generation
por: Zhong, Qihuang, et al.
Publicado: (2022)

ADEPT: Adaptive Dynamic Early-Exit Process for Transformers
por: Yoo, Sangmin, et al.
Publicado: (2026)

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents
por: Yang, Rui, et al.
Publicado: (2025)

Diversifying the Mixture-of-Experts Representation for Language Models with Orthogonal Optimizer
por: Liu, Boan, et al.
Publicado: (2023)

Improving Large Language Models with Concept-Aware Fine-Tuning
por: Chen, Michael K., et al.
Publicado: (2025)

Large Language Models as an Indirect Reasoner: Contrapositive and Contradiction for Automated Reasoning
por: Zhang, Yanfang, et al.
Publicado: (2024)

JustLogic: A Comprehensive Benchmark for Evaluating Deductive Reasoning in Large Language Models
por: Chen, Michael K., et al.
Publicado: (2025)

Better, Faster: Harnessing Self-Improvement in Large Reasoning Models
por: Zhong, Qihuang, et al.
Publicado: (2026)

Reason-KE++: Aligning the Process, Not Just the Outcome, for Faithful LLM Knowledge Editing
por: Wu, Yuchen, et al.
Publicado: (2025)

Robust Knowledge Editing via Explicit Reasoning Chains for Distractor-Resilient Multi-Hop QA
por: Wu, Yuchen, et al.
Publicado: (2025)

Revisiting Knowledge Distillation for Autoregressive Language Models
por: Zhong, Qihuang, et al.
Publicado: (2024)

Dynamic Early Exit in Reasoning Models
por: Yang, Chenxu, et al.
Publicado: (2025)

Aligning Large Language Models from Self-Reference AI Feedback with one General Principle
por: Bao, Rong, et al.
Publicado: (2024)

SpecExit: Accelerating Large Reasoning Model via Speculative Exit
por: Yang, Rubing, et al.
Publicado: (2025)

DB-LLM: Accurate Dual-Binarization for Efficient LLMs
por: Chen, Hong, et al.
Publicado: (2024)

SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token
por: Ma, Ming, et al.
Publicado: (2025)

Detoxifying Large Language Models via Autoregressive Reward Guided Representation Editing
por: Xiao, Yisong, et al.
Publicado: (2025)