AdaptArena: Evaluating Test-Time Personalization of Web Agents
cs.LG
Submitted: 2026-09-29
Updated: 2026-09-29
Code: https://github.com/McGill-NLP/web-agents-test-time-adaptations
Terminology
Sources
- Qwen3-VL Technical Report
- WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models
- Persona2Web: Benchmarking Personalized Web Agents for Contextual Reasoning with User History
- CUPID: Evaluating Personalized and Contextualized Alignment of LLMs from Interactions
- VideoAgentTrek: Computer Use Pretraining from Unlabeled Videos
- WebLINX: Real-World Website Navigation with Multi-Turn Dialogue
- JEF-Hinter: Leveraging Offline Knowledge for Improving Web Agents Adaptation
- WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning
- T-POP: Test-Time Personalization with Online Preference Feedback
- LaMP: When Large Language Models Meet Personalization
- Beyond Dialogue Time: Temporal Semantic Memory for Personalized LLM Agents
- Tent: Fully Test-time Adaptation by Entropy Minimization
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- OpenCUA: Open Foundations for Computer-Use Agents
- Agent Workflow Memory
- Inducing Programmatic Skills for Agentic Tasks
- Scaling Computer-Use Grounding via User Interface Decomposition and Synthesis
- ReAct: Synergizing Reasoning and Acting in Language Models
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- Mobile-Agent-v3: Fundamental Agents for GUI Automation
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks