SARBLOH 证明 ARC-AGI-3 的瓶颈在感知而非规划
Transforming Integer Grids to Spatial Intuition: Perception, Not Planning, Governs ARC-AGI-3
MIT 那帮人做了个 SARBLOH，只加了个感知模块就在 Kaggle 评测拿到首个冠军 13.7 倍的分数，思路很值得琢磨。
ARC-AGI-3 用未提供说明的视觉游戏考察智能体的流体智能，评分采用 Relative Human Action Efficiency，对探索浪费进行二次方惩罚。研究团队发现，前沿模型直接读取原始整数网格时，序列化破坏了空间拓扑，需要数百步动作才能重新发现基本边界和对称性。他们推出的开源智能体 SARBLOH 用一个零参数的 O(HW) 感知引擎，在 35 毫秒内将每帧转换为物体清单、几何简报和标注画布，并配合 code-as-action 运行时生成算法假设。在 23 个留存游戏上，几何不变量以 100% 准确率泛化，而单帧可供性猜测准确率仅 6-50%。在 9 小时预算、单张 Kaggle RTX PRO 6000 GPU 的条件下，SARBLOH 在隐藏评测中取得 16.53 分，达到首届 Milestone 冠军成绩的 13.7 倍。
Transforming Integer Grids to Spatial Intuition: Perception, Not Planning, Governs ARC-AGI-3
Sample-efficient fluid intelligence requires agents that rapidly form working models of unfamiliar interactive environments. ARC-AGI-3 tests this across uninstructed visual games scored by Relative Human Action Efficiency, which penalizes exploratory waste quadratically. When frontier models read raw integer grids, serialization destroys spatial topology, and they spend hundreds of actions rediscovering basic boundaries and symmetries. We introduce SARBLOH, an open-weight agent built on the premise that spatial perception, not planning capacity, governs interactive reasoning on ARC-AGI-3. A zero-parameter O(HW) perception engine turns each frame, in 35 ms, into an object inventory with bipartite temporal tracking, a geometric briefing that separates proven invariants from affordance hypotheses, and an annotated canvas. Paired with a code-as-action runtime, the agent writes algorithmic hypotheses instead of low-level moves. Across eight presentations, grounded perception raises the rubric score from 3/28 to 25/28 points while cutting token use by 89.4%. On 23 held-out games, geometric invariants generalize with 100% accuracy, while single-frame affordance guesses fall to 6-50%, so active verification probes remain necessary. On the public set, Run A (briefing off) solved 143 of 183 levels in 8.2 hours at 59.4% single-pass efficiency; Run B (briefing on) reached 63.5%. Within the nine-hour budget on one Kaggle RTX PRO 6000 GPU, SARBLOH scored 16.53 on the hidden Kaggle evaluation, 13.7x the first Milestone winner.
