paper-with-me

Papers

Beyond Perplexity: UTF-8 Validity in Byte-aware Language Models

2026-06-12 · Sangwhan Moon, Daisuke Oba, Youmi Ma, Tatsuya Hiraoka, Naoaki Okazaki arxiv

Byte-level tokenization enables language models to handle any Unicode input, but models can generate invalid UTF-8 sequences when encountering rare or unseen characters. We investigate the relationship between training scale and UTF-8 generation reliability with a 355M parameter model trained on 80B tokens from a balanced multilingual corpus of English, Japanese, Korean, and Chinese. We introduce multiple evaluation protocols that isolate UTF-8 structural validity from language modeling. UTF-8 validity convergence lags perplexity by a roughly a factor of two: perplexity stabilizes after 2.1B tokens, but UTF-8 validity requires 4.2B tokens. In context-free generation, rare characters achieve higher structural validity than common characters, suggesting over-specialization of frequent character representations. Through experiments, we observed that reliable UTF-8 generation is a distinct capability requiring evaluation beyond perplexity.

📄 PDF Abstract BibTeX arXiv:2606.14122

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Network-Efficient World Model Token Streaming

2026-05-11 · Shatadal Mishra, Ahmadreza Moradipari, Nejib Ammar arxiv

Generative driving world models rely on compact latent state representations that must be efficiently transmitted and synchronized across distributed compute and connected vehicles. We study network-efficient streaming o…

Rectify Evaluation Preference: Improving LLMs' Critique on Math Reasoning via Perplexity-aware Reinforcement Learning

2025-11-13 · Changyuan Tian, Zhicong Lu, Shuang Qian, Nayu Liu 외 arxiv

To improve Multi-step Mathematical Reasoning (MsMR) of Large Language Models (LLMs), it is crucial to obtain scalable supervision from the corpus by automatically critiquing mistakes in the reasoning process of MsMR and …

Reinforcement LearningMathematical Reasoning

Significance-Gain Pair Encoding for LLMs: A Statistical Alternative to Frequency-Based Subword Merging

2026-02-26 · Azam Nouri arxiv

Subword tokenization is a key design choice for modern language models, including large language models (LLMs), with byte- and character-level BPE serving as a widely used baseline. Standard BPE selects merges by raw pai…

Can Perplexity Predict Fine-Tuning Performance? An Investigation of Tokenization Effects on Sequential Language Models for Nepali

2024-04-28 · Nishant Luitel, Nirajan Bekoju, Anand Kumar Sah, Subarna Shakya

Recent language models use subwording mechanisms to handle Out-of-Vocabulary(OOV) words seen during test time and, their generation capacity is generally measured using perplexity, an intrinsic metric. It is known that i…

Language ModelingLanguage Modelling

Equity with Efficiency: An Empirical Study of Tokenizers for Multilingual Large Language Models

2026-06-13 · Kieron Seven Jun Wei Lee, Muhammad Reza Qorib, Andrew Ivan Soegeng, Hwee Tou Ng arxiv

Multilingual large language models (LLMs) depend on subword tokenization to bridge discrete text and continuous neural representation. State-of-the-art multilingual LLMs often use Byte-level Byte-Pair Encoding (BPE) toke…