Tag
This paper introduces an audit that tests whether LLM pricing agents trained with RL actually enforce collusion by punishing cheaters, finding that small RL-trained LLMs mostly learn fixed rules or deterrence rather than true cartel enforcement, and that outcome measures alone cannot distinguish genuine collusion from rule-following.
The paper proposes ReLOBGen, a limit order book message generator that ensures replayability by construction by selecting referenced orders from the current resting book and masking invalid tokens, achieving 100% replayability in 500-message rollouts and a 2.7-3.6x speedup per replayed message over the LOBS5 baseline.
MarketFish is an open-source multi-agent market simulation engine that generates 128+ AI consumers with distinct identities, budgets, and biases to simulate purchase decisions over 30 rounds, helping founders validate product ideas before launch.
Anthropic conducted an experiment where employees used Claude-powered agents to trade books in a miniature market, studying agent representation and how model strength affects negotiating outcomes.