Personalized Help for Optimizing Low-Skilled Users' Strategy

Research Summary

Contribution Summary

• AIs can beat humans in game environments; however, how helpful those agents are to human remains understudied.
• We augment CICERO, a natural language agent that demonstrates superhuman performance in Diplomacy, to generate both move and message advice based on player intentions.
• A dozen Diplomacy games with novice and experienced players, with varying advice settings, show that some of the generated advice is beneficial.

Why It Matters For Eval

• AIs can beat humans in game environments; however, how helpful those agents are to human remains understudied.
• We augment CICERO, a natural language agent that demonstrates superhuman performance in Diplomacy, to generate both move and message advice based on player intentions.

Related Papers

• DCoPilot: Generative AI-Empowered Policy Adaptation for Dynamic Data Center Operations Embedding Neighbor
• Diagnosing LLM Reranker Behavior Under Fixed Evidence Pools Embedding Neighbor
• HoloLLM: Multisensory Foundation Model for Language-Grounded Human Sensing and Reasoning Embedding Neighbor
• Luna-2: Scalable Single-Token Evaluation with Small Language Models Embedding Neighbor
• Towards Controllable Video Synthesis of Routine and Rare OR Events Embedding Neighbor
• Probabilistic distances-based hallucination detection in LLMs with RAG Embedding Neighbor
• AgenticRAGTracer: A Hop-Aware Benchmark for Diagnosing Multi-Step Retrieval Reasoning in Agentic RAG Embedding Neighbor
• Embedding Retrofitting: Data Engineering for better RAG Embedding Neighbor
• Language and Experience: A Computational Model of Social Learning in Complex Tasks Tag Overlap
• Towards interpretable models for language proficiency assessment: Predicting the CEFR level of Estonian learner texts Tag Overlap
• RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environments Tag Overlap
• EconEvals: Benchmarks and Litmus Tests for Economic Decision-Making by LLM Agents Tag Overlap