Scratchpad Patching: Decoupling Compute from Patch Size in Byte-Level Language Models
About
Tokenizer-free language models eliminate the tokenizer step of the language modeling pipeline by operating directly on bytes; patch-based variants further aggregate contiguous byte spans into patches for efficiency. However, the average patch size chosen at the model design stage governs a tight trade-off: larger patches reduce compute and KV-cache footprint, but degrade modeling quality. We trace this trade-off to patch lag: until a patch is fully observed, byte predictions within it must rely on a stale representation from the previous patch to preserve causality; this lag widens as patches grow larger. We introduce Scratchpad Patching (SP), which inserts transient scratchpads inside each patch to aggregate the bytes seen so far and refresh patch-level context for subsequent predictions. SP triggers scratchpads using next-byte prediction entropy, selectively allocating compute to information-dense regions and enabling post-hoc adjustment of inference-time compute. Across experiments on natural language and code, SP improves model quality at the same patch size; for example, even at $16$ bytes per patch, SP-augmented models match or closely approach the byte-level baseline on downstream evaluations while using a $16\times$ smaller KV cache over patches and $3$-$4\times$ less inference compute.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Language Understanding | MMLU | Accuracy34.7 | 844 | |
| Question Answering | OpenBookQA | Accuracy48 | 305 | |
| Question Answering | BoolQ | Accuracy66.3 | 201 | |
| Natural Language Understanding | ARC Easy | Accuracy71 | 36 | |
| Natural Language Understanding | HellaSwag | Accuracy59.7 | 35 | |
| Natural Language Understanding | ARC-C | Accuracy41.9 | 34 | |
| Natural Language Understanding | WinoGrande | Accuracy59 | 30 | |
| Natural Language Understanding | PIQA | PIQA Accuracy73.8 | 16 |