Jul 2026· Signal Processing and Communications Applications Conference· pp. 1-4· 0 citations· 9 references
Abstract
This paper quantifies the context tax imposed by general-purpose tokenization in agglutinative languages. Using the same local Byte Pair Encoding (BPE) recipe, we train 64k native tokenizers for Turkish, Finnish, and Hungarian and compare them with tokenizers of GPT-4, mBERT, and XLM-R models. Each native tokenizer is trained on 30,000 Wiki40B texts and evaluated on a disjoint 500-text test set. The main metric is fertility, the average number of subword tokens per word; we also convert it into approximate word coverage under a fixed 512-token budget. In all three languages, the native tokenizer produces the shortest sequences. Native-vs-GPT-4 fertility gains are 77.63% for Turkish, 64.42% for Finnish, and 79.21% for Hungarian. The results show that tokenizer choice is a first-order context-efficiency decision in agglutinative languages.
Large language models (LLMs) process text through subword tokenizers rather than directly reading characters or words. Because these tokenizers are trained predominantly on English-centric corpora, they introduce a systematic and often overlooked disadvantage for many non-English languages. In this work, we quantify this tokenizer tax for Indian languages using the FLORES-200 parallel corpus, measuring tokenization fertility across six widely used tokenizers and fourteen languages. Under cl100k_base (used by GPT-3.5 and GPT-4), Indian languages experience an average 8.0x tokenization tax relative to English, reaching 13.0x for Malayalam, reducing the effective context window to as little as 12% of that available to English users for equivalent semantic content. We identify the primary mechanism behind this disparity: failed byte-pair merges that leave text fragmented into single-byte tokens, with merge failure strongly correlating with tokenizer tax (Pearson r = 0.89). We further show that this phenomenon is not an inherent property of Indic scripts but a consequence of tokenizer design. Multilingual tokenizers such as XLM-R and OpenAI's o200k_base reduce the average Indic tokenizer tax by 73%, demonstrating that the disparity is largely remediable. Beyond token statistics, we quantify a practical consequence by showing that, under fixed context budgets, Indian-language documents preserve substantially less original content than equivalent English documents. Finally, we examine the relationship between tokenizer fertility and reading comprehension performance on the Belebele benchmark, finding that the apparent correlation is largely explained by language resource availability rather than tokenizer behavior alone.
Byte-level BPE tokenizers that use the HuggingFace ByteLevel pre-tokenizer inherit GPT-2's word regex, where a word is defined as \p{L}+, one or more Unicode letters. In abugida scripts, vowels are written as combining marks; this pattern therefore splits each word at every vowel sign. Since BPE merges only within a pre-token, those splits persist through training regardless of vocabulary size or corpus composition. We formalise this effect as a training-free lower bound on fertility. Across 26 languages from a parallel corpus, every one of the 17 abugidas is affected, ranging from 1.47x (Tibetan) to 9.02x (Thai), whereas Latin, Cyrillic, Hangul, and Han show exactly 1.00x. For 5 languages, matched tokenizer pairs that differ only in this character class fall within 2.2% of the predicted floor, scoring 4.78 versus 1.58 tokens per word on Nepali. When the Nepali share of the training corpus is swept from 5% to 95%, the broken tokenizer barely shifts at all (1.7%) while the fixed one shifts 33.9%, which separates a structural ceiling from a data shortage without needing to inspect any code. We train three 268M models that differ only in their tokenizer; the fixed variant achieves 4.43% lower held-out Nepali bits per byte at equal compute, and it still leads when given the same bytes with 1.59x the compute. A census of 3,479 HuggingFace repositories finds the letters-only word class present in 63.3% of the most-downloaded text-generation models, accounting for 72.5% of their downloads. GPT-4o's o200k pattern already uses a mark-aware word class, making the repair itself prior art. We quantify its value, show how to recognise its absence from symptoms alone, map which scripts it reaches, measure how widely it is deployed, and release a 65,536-entry Nepali-English tokenizer with a harness that regenerates every number here from public data on a laptop.
S. Regmi, Siddhartha Pudasaini, Chetan Phakami Pun· 1 citation
Large language models pay a well-documented tax on non-English text: the same content costs several times more tokens, and because attention is quadratic in sequence length, far more compute. We ask how much of this tax is removable. Framing the token layer as source coding -- transformer compute is monotone in sequence length, whose per-atom floor is the Shannon rate $H/\log_2 V$, an object already applied to tokenizers in prior work -- we assemble a token-cost ledger that splits each language's cost, at fixed parallel content, into a removable coding redundancy, a residual coding slack, an intrinsic-content term, and an orthogonal, irreducible grapheme-to-phoneme term that governs the multimodal rather than the text cost. On FLORES-200 across eight languages, a production tokenizer costs up to $8.9\times$ more tokens for Indic scripts than for English; a script-matched code trained on $1,012$ sentences removes a median $64\%$ of that excess (bootstrap 95\% CI $[0.638, 0.647]$), and a script-fair information floor shows the intrinsic content differs by under $6\%$ -- the tax is representational, not informational. A constructed code removes $98\%$ of a controlled source's redundancy, and the token tax implies up to $79\times$ attention cost. We are explicit about scope and failure: this is compute-and-memory accounting, not a model-quality claim; we neither measure nor claim the cross-lingual direction of the orthographic term; and our matched code is a conservative small-data demonstration. We contribute the unifying ledger, the removable-versus-intrinsic attribution, and an open one-command harness.
Modern multilingual tokenizers often fragment Ukrainian and other underrepresented Cyrillic-script languages more heavily than English, creating disparities in cost and context capacity. We quantify this overhead across nine production tokenizers and five languages with standardized Cyrillic and Latin representations, covering 8.37 million word forms. On a corpus benchmark, Ukrainian shows 68-121% token overhead on modern tokenizers and 220% on the older cl100k, measured through full-text fertility on the BrUK and Brown corpora. Overhead is negatively associated with Cyrillic vocabulary allocation in the subset with independently verified English baselines, although the association is not statistically significant (Spearman rho = -0.536, p = 0.215, n = 7). We evaluate two mitigation strategies. LLMLingua-2 reduces Ukrainian input length by 47-49% on an e-commerce RAG benchmark of 1,536 products and 145 queries, with no compression-induced value losses among 80 retrievable cases. A balanced byte-level BPE tokenizer trained with a 200K vocabulary cap, converging at 158,184 actual entries, reduces the held-out UK/EN ratio from 2.22x to 1.30x. Romanization increases Ukrainian token counts by 2-19% on most tokenizers. Across the five languages, tokenization efficiency favors the script more prevalent in web data. These findings indicate that training data allocation contributes to Cyrillic tokenization overhead and that mitigation is possible at both inference and tokenizer-design stages.
A tokenizer fixed at the start of pre-training allocates vocabulary in proportion to the pre-training corpus, reflecting the deployment priorities at that time. When those priorities shift, languages added later are split into many more tokens per word, which can raise latency, compute, and energy consumption for users of those languages. Cloud models can afford a broad vocabulary because the embedding and LM-head matrices are a small fraction of their parameters. On a compact model those matrices are a material share of per-token decode bandwidth, so on-device models ship small vocabularies and accept fragmentation outside a fixed language set. We present tokenizer expansion, an in-place recipe for upgrading a pre-trained model's tokenizer when the model producer controls its design. We continue the existing tokenizer's BPE merges on a multilingual corpus, so most source tokens carry over unchanged as single tokens and every new token has an exact decomposition into source tokens. We copy the carried-over embedding rows unchanged and initialize new rows as the mean of their source sub-token embeddings. A two-stage adaptation, embedding-only training then full-model continued pre-training, recovers source-checkpoint quality. We apply the recipe to a continued pre-trained checkpoint of LFM2-8B-A1B, an 8B-parameter Mixture-of-Experts model, to help produce LFM2.5-8B-A1B with a 128K tokenizer. The expanded tokenizer encodes Hindi and Vietnamese in roughly $2.4\times$ and $2.6\times$ fewer tokens than the source (up to $4.0\times$ on Thai). Combining these reductions with the measured per-token cost of the larger vocabulary, we estimate a $2.2$-$3.7\times$ per-character decode speedup for these languages across our reference devices. We release the model weights and the expanded tokenizer, and report the negative findings that shaped the recipe.
Jimmy T.H. Smith, Tarek Dakhran, Alberto Cabrera et al.· 0 citations
The experimental results revealed the strengths and weaknesses of various approaches to subword segmentation and identified the most effective tokenization strategies under the conditions of the morphological complexity of the Tajik language.
M. Arabov, S. Khaybullina· Automatic Documentation and...· 0 citations