Search papers, labs, and topics across Lattice.
NVIDIAAffiliation:IIT DelhiIndependent ResearcherNTUSingapore Institute of TechnologyAug 3, 2026arXiv:2608.02050This paper introduces TextNCA, a 1D causal windowed-attention model based on the Neural Cellular Automaton (NCA) framework, exploring its effectiveness for language modeling. The authors conduct experiments on WikiText-103, revealing that while TextNCA does not outperform a parameter-matched Transformer, it serves as an analytical tool to understand the impact of hierarchical local attention and iteration on performance. Key findings indicate that the model's behavior is heavily influenced by the staged narrow-to-wide attention schedule, with specific configurations yielding significant variations in perplexity scores.
Hierarchical local attention in TextNCA reveals that the arrangement of attention windows can dramatically influence language modeling performance, even more than the model's iterative nature.
Can a strictly local, iterated, weight-shared computation primitive support language modelling, and which of those three properties actually drives the model's behaviour? We define \textsc{TextNCA}, a 1D causal windowed-attention realisation of the Neural Cellular Automaton primitive, and study a hierarchical variant that cascades three stages with windows $w \in \{8, 32, 128\}$ and $T_s$ shared-weight iterations per stage, all on WikiText-103 at roughly 30M parameters and 60k training steps. The model does not match a parameter-matched Transformer at this scale (Hier-TextNCA $60.3$ vs.\ Transformer-6L $52.8$ and Transformer-12L $44.7$ PPL), so we treat it as an analytical probe rather than a proposed alternative. The behaviour we observe is largely explained by the staged narrow-to-wide schedule: a non-iterating sliding-window Transformer that reuses the same schedule comes within $+4.1$ PPL of the iterated model, while reversing, flattening, or breaking the monotonic ordering of the schedule costs between $+16.7$ and $+70.8$ PPL. Iteration adds a smaller bounded benefit on top of the schedule, with a clear optimum at $T_s{=}4$ and a U-shaped degradation beyond it. The GRU gate and learned per-step embeddings are required for that benefit to appear, and training with random $T_s$ yields an inference-time iteration-count knob at the cost of substantially higher absolute PPL. We position the work as a controlled reading of which parts of NCA-style computation carry the weight in language modelling.
