From 7e7e6a51d3137a4e1319b841ad6c397f6d65c468 Mon Sep 17 00:00:00 2001 From: Jack Lanchantin Date: Mon, 30 Mar 2026 09:42:08 -0400 Subject: [PATCH] Update inference section with figure reference Added reference to inference figure in the inference section. --- projects/paragator/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/projects/paragator/README.md b/projects/paragator/README.md index 7977785..0342337 100644 --- a/projects/paragator/README.md +++ b/projects/paragator/README.md @@ -107,7 +107,7 @@ Intuitively, the model is rewarded when at least one of its $m$ attempts solves ### Inference -During training, we optimize a single round of aggregation over one candidate pool. At inference time, however, we naturally generalize this to multiple iterations of aggregation, enabling sequential scaling in addition to the learned parallel sampling. +During training, we optimize a single round of aggregation over one candidate pool. At inference time, however, we naturally generalize this to multiple iterations of aggregation, enabling sequential scaling in addition to the learned parallel sampling (see inference figure above). Concretely, given problem $x$, we first sample an initial pool of $m$ candidates $y_{1:m}^{(0)} \sim \mathcal{M}_\theta(y \mid x)$ and sample $m$ aggregated solutions