In controlled experiments, 2D positional encoding outperforms 1D, RoPE, and learned embeddings for transformer models on ARC-like tasks when training examples are limited.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
The role of positional encodings in the ARC benchmark
In controlled experiments, 2D positional encoding outperforms 1D, RoPE, and learned embeddings for transformer models on ARC-like tasks when training examples are limited.