Multi-agent Evaluation

Strategic bargaining agent combining LLM reasoning with deterministic constraint enforcement. Uses GPT-4o-mini to propose allocations and enforces M4/M5 rules to avoid accepting offers below BATNA or walking away from profitable deals.

Agent that negotiates in multi-round bargaining games using LLM reasoning. Evaluated on MAizeBargAIn benchmark.

Multi-round bargaining agent for the MAizeBargAIn meta-game assessor. Combines LLM reasoning (Gemini 2.5 Flash-Lite) with a deterministic M1–M5 rule validator for guaranteed feasible actions.

Test

This green agent evaluates single-cell perturbation significance analysis as a binary QA task. Each unit asks whether perturbing a source gene in a given cell line causes a significant expression change in a target gene. The participant must answer strictly in the format “Final Answer: Yes/No”.