X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization
cs.AI
Submitted: 2026-09-26
Updated: 2026-09-26
Comments: Project in Progress. Homepage: https://sitaocheng.github.io/xtree/. Code: https://github.com/sitaocheng/X-Tree/
Code: https://github.com/sitaocheng/X-Tree
Project page: https://sitaocheng.github.io/xtree
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Thread: A Logic-Based Data Organization Paradigm for How-To Question Answering with Retrieval Augmented Generation
- Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models
- Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments
- Dual-Evolving Reinforcement Learning for Automated Reward Design
- Atomic Skills are the Prerequisite: When Reinforcement Learning Synthesizes Compositional Reasoning, and When It Only Amplifies
- Group-in-Group Policy Optimization for LLM Agent Training
- Multi-Level Discovery of Deep Options
- Go-Browse: Training Web Agents with Structured Exploration
- QueryAgent: A Reliable and Efficient Reasoning Framework with Environmental Feedback-based Self-Correction
- TARGA: Targeted Synthetic Data Generation for Practical Reasoning over Structured Data
- Disentangling Memory and Reasoning Ability in Large Language Models
- Self-Distilled Agentic Reinforcement Learning
- WALT: Web Agents that Learn Tools
- WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning
- Neural Machine Translation of Rare Words with Subword Units
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Self-Distillation Enables Continual Learning
- Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents
- Learn-by-interact: A Data-Centric Framework for Self-Adaptive Agents in Realistic Environments
- Voyager: An Open-Ended Embodied Agent with Large Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems