Imagen de Portada

Just Do It!? Computer-Use Agents Exhibit Blind Goal-Directedness

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Shayegani, Erfan, Hines, Keegan, Dong, Yue, Abu-Ghazaleh, Nael, Lutz, Roman, Whitehead, Spencer, Balachandran, Vidhisha, Nushi, Besmira, Vineet, Vibhav
Formato:	Preprint
Publicado:	2025
Materias:	Artificial Intelligence Computation and Language Cryptography and Security Computers and Society Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

MM-GEN: Enhancing Task Performance Through Targeted Multimodal Data Curation
por: Joshi, Siddharth, et al.
Publicado: (2025)

Modeling Hierarchical Thinking in Large Reasoning Models
por: Shahariar, G M, et al.
Publicado: (2025)

Misaligned Roles, Misplaced Images: Structural Input Perturbations Expose Multimodal Alignment Blind Spots
por: Shayegani, Erfan, et al.
Publicado: (2025)

Improving Instruction-Following in Language Models through Activation Steering
por: Stolfo, Alessandro, et al.
Publicado: (2024)

BenchAgents: Multi-Agent Systems for Structured Benchmark Creation
por: Butt, Natasha, et al.
Publicado: (2024)

Eureka: Evaluating and Understanding Large Foundation Models
por: Balachandran, Vidhisha, et al.
Publicado: (2024)

Unearthing Skill-Level Insights for Understanding Trade-Offs of Foundation Models
por: Moayeri, Mazda, et al.
Publicado: (2024)

Inference-Time Scaling for Complex Tasks: Where We Stand and What Lies Ahead
por: Balachandran, Vidhisha, et al.
Publicado: (2025)

VLMs Need Words: Vision Language Models Ignore Visual Detail In Favor of Semantic Anchors
por: Shahgir, Haz Sameen, et al.
Publicado: (2026)

Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning
por: Vilas, Martina G., et al.
Publicado: (2025)

Layer-wise Alignment: Examining Safety Alignment Across Image Encoder Layers in Vision Language Models
por: Bachu, Saketh, et al.
Publicado: (2024)

Cross-Modal Safety Alignment: Is textual unlearning all you need?
por: Chakraborty, Trishna, et al.
Publicado: (2024)

Attention Speaks Volumes: Localizing and Mitigating Bias in Language Models
por: Adiga, Rishabh, et al.
Publicado: (2024)

That Doesn't Go There: Attacks on Shared State in Multi-User Augmented Reality Applications
por: Slocum, Carter, et al.
Publicado: (2023)

Evil Vizier: Vulnerabilities of LLM-Integrated XR Systems
por: Zhang, Yicheng, et al.
Publicado: (2025)

GPUVM: GPU-driven Unified Virtual Memory
por: Nazaraliyev, Nurlan, et al.
Publicado: (2024)

Poison Once, Refuse Forever: Weaponizing Alignment for Injecting Bias in LLMs
por: Mamun, Md Abdullah Al, et al.
Publicado: (2025)

Evaluating the Goal-Directedness of Large Language Models
por: Everitt, Tom, et al.
Publicado: (2025)

A Behavioural and Representational Evaluation of Goal-Directedness in Language Model Agents
por: Arghal, Raghu, et al.
Publicado: (2026)

Physics Knowledge in Frontier Models: A Diagnostic Study of Failure Modes
por: Bagdonaviciute, Ieva, et al.
Publicado: (2025)

Co(ve)rtex: ML Models as storage channels and their (mis-)applications
por: Mamun, Md Abdullah Al, et al.
Publicado: (2023)

Diversity of Thought Improves Reasoning Abilities of LLMs
por: Naik, Ranjita, et al.
Publicado: (2023)

Elephants Never Forget: Memorization and Learning of Tabular Data in Large Language Models
por: Bordt, Sebastian, et al.
Publicado: (2024)

Detecting Data Contamination in LLMs via In-Context Learning
por: Zawalski, Michał, et al.
Publicado: (2025)

Fara-7B: An Efficient Agentic Model for Computer Use
por: Awadallah, Ahmed, et al.
Publicado: (2025)

From Measurement to Expertise: Empathetic Expert Adapters for Context-Based Empathy in Conversational AI Agents
por: Shayegani, Erfan, et al.
Publicado: (2025)

Understanding Information Storage and Transfer in Multi-modal Large Language Models
por: Basu, Samyadeep, et al.
Publicado: (2024)

Phi-4-reasoning Technical Report
por: Abdin, Marah, et al.
Publicado: (2025)

Navigating Hallucinations for Reasoning of Unintentional Activities
por: Grover, Shresth, et al.
Publicado: (2024)

On Occlusions in Video Action Detection: Benchmark Datasets And Training Recipes
por: Modi, Rajat, et al.
Publicado: (2024)

Exposing the Achilles' Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical Reasoning
por: Singh, Joykirat, et al.
Publicado: (2024)

ReVision: Scaling Computer-Use Agents via Temporal Visual Redundancy Reduction
por: Abaskohi, Amirhossein, et al.
Publicado: (2026)

HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video Understanding
por: Azad, Shehreen, et al.
Publicado: (2025)

StreamReady: Learning What to Answer and When in Long Streaming Videos
por: Azad, Shehreen, et al.
Publicado: (2026)

Measuring Goal-Directedness
por: MacDermott, Matt, et al.
Publicado: (2024)

A Large-Scale Analysis on Contextual Self-Supervised Video Representation Learning
por: Kumar, Akash, et al.
Publicado: (2025)

OmViD: Omni-supervised active learning for video action detection
por: Rana, Aayush, et al.
Publicado: (2025)

Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration
por: Feng, Shangbin, et al.
Publicado: (2024)

Knowledge Card: Filling LLMs' Knowledge Gaps with Plug-in Specialized Language Models
por: Feng, Shangbin, et al.
Publicado: (2023)

Reasoning Up the Instruction Ladder for Controllable Language Models
por: Zheng, Zishuo, et al.
Publicado: (2025)