← 研踪 PaperTrailarXiv:2609.22308 · v2
GameReplica: A Benchmark for Black-Box Visual Game Replication by Vision-Language Agents
首次提交 2026-09-15 11:07
版本更新 2026-11-16 12:47
雷达观测 2026-11-17 17:12
摘要
Benchmarks for coding agents typically assess implementation from explicit specifications, leaving open whether agents can discover and reproduce systems whose rules are undocumented and accessible only through interaction. We study this capability in web games, which offer controlled environments with observable actions and outcomes, and introduce GameReplica, a benchmark for end-to-end black-box game replication. It comprises 125 instances spanning 25 web games and five core mechanism families, with each game instantiated at five difficulty levels. Under container isolation, an agent accesses the target game only through an action--screenshot interface, may iterate between probing the game, implementing a replica, and checking it, and delivers a self-contained runnable replica together with a rule checklist. A reference-grounded evaluation then scores the deliverables separately on rule consistency, implementation consistency, and visual fidelity. Experiments across six models and five agent runtimes show that black-box replication remains challenging: Claude Opus 4.8 achieves the highest Overall score, 71.6%, while the other models obtain 4.0%--43.0%. Every model scores highest on visual fidelity and lowest on rule consistency, suggesting that reproducing appearance is easier than recovering mechanics. Difficulty widens the gap: from L1 to L5, the four lower-scoring models show relative Overall score declines of 33%--63%, whereas the strongest model declines by less than 5%. In the evaluated configurations, outcomes vary more across models than across agent runtimes, and most attributed failures of weaker models arise during exploration, before replica code is written. Resources and interactive demonstrations are available at https://qqqqqqby.github.io/GameReplica/.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-16 12:47 · 最新观测版本
GameReplica: A Benchmark for Black-Box Visual Game Replication by Vision-Language Agents
Benchmarks for coding agents typically assess implementation from explicit specifications, leaving open whether agents can discover and reproduce systems whose rules are undocumented and accessible only through interaction. We study this capability in web games, which offer controlled environments with observable actions and outcomes, and introduce GameReplica, a benchmark for end-to-end black-box game replication. It comprises 125 instances spanning 25 web games and five core mechanism families, with each game instantiated at five difficulty levels. Under container isolation, an agent accesses the target game only through an action--screenshot interface, may iterate between probing the game, implementing a replica, and checking it, and delivers a self-contained runnable replica together with a rule checklist. A reference-grounded evaluation then scores the deliverables separately on rule consistency, implementation consistency, and visual fidelity. Experiments across six models and five agent runtimes show that black-box replication remains challenging: Claude Opus 4.8 achieves the highest Overall score, 71.6%, while the other models obtain 4.0%--43.0%. Every model scores highest on visual fidelity and lowest on rule consistency, suggesting that reproducing appearance is easier than recovering mechanics. Difficulty widens the gap: from L1 to L5, the four lower-scoring models show relative Overall score declines of 33%--63%, whereas the strongest model declines by less than 5%. In the evaluated configurations, outcomes vary more across models than across agent runtimes, and most attributed failures of weaker models arise during exploration, before replica code is written. Resources and interactive demonstrations are available at https://qqqqqqby.github.io/GameReplica/.
查看此版本解读 →查看此版本原文 ↗