Tomas Pfister

CoDA: Agentic Systems for Collaborative Data Visualization

ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory

ATLAS: Constraints-Aware Multi-Agent Collaboration for Real-World Travel Planning

TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture

Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning

DS-STAR: Data Science Agent via Iterative Planning and Verification

Watch and Learn: Learning to Use Computers from Online Videos

LLM-based Multi-Agent Blackboard System for Information Discovery in Data Science

HEART: Emotionally-driven test-time scaling of Language Models

VeriGuard: Enhancing LLM Agent Safety via Verified Code Generation

Cost-effective Agent Test-time Scaling via Budget-Aware Thinking

MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement

Model Swarms: Collaborative Search to Adapt LLM Experts via Swarm Intelligence

Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems

Model Swarms: Collaborative Search to Adapt LLM Experts via Swarm Intelligence

Mitigating Object Hallucination in MLLMs via Data-augmented Phrase-level Alignment

Towards Compute-Optimal Many-Shot In-Context Learning

Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training

Speculative Knowledge Distillation: Bridging the Teacher-Student Gap Through Interleaved Sampling

SQL-GEN: Bridging the Dialect Gap for Text-to-SQL Via Synthetic Data And Model Merging

Speculative RAG: Enhancing Retrieval Augmented Generation through Drafting