lechmazur/step_game

View on GitHub ↗ ★ 93

Multi-Agent Step Race Benchmark: Assessing LLM Collaboration and Deception Under Pressure. A multi-player “step-race” that challenges LLMs to engage in public conversation before secretly picking a move (1, 3, or 5 steps). Whenever two or more players choose the same number, all colliding players fail to advance.

7 mentions on Hacker News by 2 people: 2 stories, 5 comments · first 2025-01-22 · latest 2026-01-21

Mentions per year 7 mentions · busiest year 2025 (6)

20082026

All Stories Comments Newest Top Oldest

…- Step Race Benchmark: Assessing LLM Collaboration and Deception Under Pressure at https://github.com/lechmazur/step_game/
comment by zone411 · 2026-01-21 · in Which AI Lies Best? A game theory classic designed by John Nash
…are very interesting to examine. I have two simple ones in another multi-agent benchmark, https://github.com/lechmazur/step_game (SilentGreedyPlayer and SilentRandomPlayer), and it's fascinating to see LLMs detect and respond to them. The…
comment by zone411 · 2025-04-07 · in Benchmarking LLM social skills with an elimination game
…planning to create game versions of this benchmark, as well as my other multi-agent benchmarks ( https://github.com/lechmazur/step_game , https://github.com/lechmazur/pgg_bench/ , and a few others I'm developing). But I'm not sure if a…
comment by zone411 · 2025-04-07 · in Benchmarking LLM social skills with an elimination game
…deception benchmarks within a couple of days. ( https://github.com/lechmazur/elimination_game/ , https://github.com/lechmazur/step_game and https://github.com/lechmazur/goods ).
comment by zone411 · 2025-02-28 · in GPT-4.5
…LLM Confabulation (Hallucination): https://github.com/lechmazur/confabulations/ LLM Step Game: https://github.com/lechmazur/step_game LLM Thematic Generalization Benchmark: https://github.com/lechmazur/generalization LLM Creative…
comment by zone411 · 2025-02-10 · in PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models