monte carlo tree search
A search algorithm that uses random sampling to estimate the value of moves in decision-making processes, particularly in games. It builds a search tree during play and uses the results of random simulations to inform its strategy, balancing exploration and exploitation effectively.
- AutoDiscovery: Open-ended Scientific Discovery via Bayesian Surprise
- Bag of Tricks for Inference-time Computation of LLM Reasoning
- Diffusion Tree Sampling: Scalable inference‑time alignment of diffusion models
- Feedback-Aware MCTS for Goal-Oriented Information Seeking
- HumanoidGen: Data Generation for Bimanual Dexterous Manipulation via LLM Reasoning
- Improving Monte Carlo Tree Search for Symbolic Regression
- MALinZero: Efficient Low-Dimensional Search for Mastering Complex Multi-Agent Planning
- PlanU: Large Language Model Reasoning through Planning under Uncertainty
- REASONING COMPILER: LLM-Guided Optimizations for Efficient Model Serving
- RF-Agent: Automated Reward Function Design via Language Agent Tree Search
- SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agents
- SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents
- SIGMA: Refining Large Language Model Reasoning via Sibling-Guided Monte Carlo Augmentation
- SYMPHONY: Synergistic Multi-agent Planning with Heterogeneous Language Model Assembly
- SoTA with Less: MCTS-Guided Sample Selection for Data-Efficient Visual Reasoning Self-Improvement
- SolverLLM: Leveraging Test-Time Scaling for Optimization Problem via LLM-Guided Search
- Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs
- Test-Time Scaling of Diffusion Models via Noise Trajectory Search
- Uncertainty-Guided Exploration for Efficient AlphaZero Training