Skip to yearly menu bar Skip to main content


Workshop Fri, Oct 9, 2026 • 8:30 AM – 6:00 PM PDT Continental Ballroom 1

Second Tokenization Workshop

Tomasz Limisiewicz ⋅ Valentin Hofmann ⋅ Sachin Kumar ⋅ Jindřich Libovický ⋅ Jindřich Helcl ⋅ Orevaoghene Ahia ⋅ Elizabeth Salesky ⋅ Yuval Pinter ⋅ Yuki M Asano
Project Page

Abstract

Chat is not available.

Schedule

Time zone: Conference local time zone
9:00 AM
Opening
10:00 AM
Coffee Break
11:00 AM
Coffee Break
→ Dynamic Multi-Byte Prediction With Hierarchical Language Models
Abraham Owodunni ⋅ Chibuzor Okocha ⋅ Christan Grant ⋅ Tomasz Limisiewicz ⋅ Sachin Kumar
→ Beyond Frequency: Output-Vocabulary Pruning by Log-Mass Maximization
Sergei Ivanov ⋅ Egor Krasheninnikov ⋅ Nikita Kozodoi ⋅ Natalia Tarasova ⋅ Aoyu Zhang
→ Tokenisation via Convex Relaxations
Jan Tempus ⋅ Philip Whittington ⋅ Craig Schmidt ⋅ Dennis Komm ⋅ Tiago Pimentel
→ Compute Optimal Tokenization
Tomasz Limisiewicz ⋅ Artidoro Pagnoni ⋅ Srini Iyer ⋅ Mike Lewis ⋅ Sachin Mehta ⋅ Alisa Liu ⋅ Margaret Li ⋅ Gargi Ghosh ⋅ Luke Zettlemoyer
→ Defragmenting Language Models: An Interpretability-based Approach for Vocabulary Expansion
Maitrey Mehta ⋅ Nishant Subramani ⋅ Zhichao Xu ⋅ Ashim Gupta ⋅ Vivek Srikumar
→ Tokenizing Numerical and Embedding Features for LLM RecSys
Zhe Xu ⋅ Ankit Peshin ⋅ Chiyu Zhang ⋅ Feng Qi ⋅ TIK O LUI ⋅ Anil Ramakrishna ⋅ Justin Johnson ⋅ Carl Hu ⋅ Kaushik Rangadurai ⋅ Luke Simon
→ Reading a Word Two Ways: A Learned Gate over Tokenization Layouts for Frozen Language Models
Anuja Dawane ⋅ Harsha Ponnada ⋅ V Subrahmany Raghu Ram Kishore Parupudi
→ Byte-Level Tokenizer Transfer with Auxiliary Objectives
Gianluca Vico ⋅ Jindřich Libovický
→ Shorthand for Thought: Compressing LLM Reasoning via Entropy-Guided Supertokens
Zhenyu Zhao ⋅ Sander Land ⋅ Daniel M Bikel ⋅ Waseem Alshikh
→ A Pilot Study of Autocompleting Tokenizers
Samuel Wexler ⋅ Mark Hopkins
→ Model-Aware Tokenizer Transfer
Mykola Haltiuk ⋅ Aleksander Smywiński-Pohl
→ LangMAP: A Language-Adaptive Approach to Tokenization
Clara Meister ⋅ Suchir Salhan ⋅ Andrzej Szablewski ⋅ Pietro Lesci ⋅ Paula Buttery ⋅ Tiago Pimentel
→ Transducing Language Models
Vésteinn Snæbjarnarson ⋅ Samuel Kiegeland ⋅ Tianyu Liu ⋅ Reda Boumasmoud ⋅ Ryan Cotterell ⋅ Tim Vieira
→ Byte $n$-gram Encoding
Marius Tomek ⋅ Philip Whittington ⋅ Violeta Kastreva ⋅ Clara Meister ⋅ Dennis Komm ⋅ Tiago Pimentel
→ BBT: BPE-Guided Byte Transformer
Hyunho Kim ⋅ SeungYeol Baek
12:30 PM
Lunch Break
2:30 PM
Coffee Break
3:30 PM
Coffee Break
→ Joint Optimization for Greedy Longest-match Tokenization
Adhiraj Singh ⋅ Deepanshu Mody ⋅ Ghina Al Shdaifat ⋅ Hamza Alshamy ⋅ Adam Wiemerslage ⋅ Varshini Reddy ⋅ Craig Schmidt
→ When Tokenization is Secretly Output Supervision
Tanja Baeumel ⋅ Josef v Genabith ⋅ Simon Ostermann
→ Objective vs. Search: Decomposing What Makes a Good Tokeniser
Ahmetcan Yavuz ⋅ Clara Meister ⋅ Tiago Pimentel
→ Less Is More: Reducing Token Counts Without Compromising Performance
Gyeongje Cho ⋅ Yeonkyoung So ⋅ Sangmin Lee ⋅ Jaejin Lee
→ What Do We Talk About When We Talk About Tokenization?
Gianluca Vico ⋅ Jindřich Libovický
→ Comparative Study of Tokenizer Replacement, Embedding Initialization, and Continued Pretraining for LLM Language Adaptation
Aleksandra Krasnodębska ⋅ Mikołaj Pokrywka ⋅ Marek Jeliński ⋅ Antoni Lasik ⋅ Marcin Bagnowski ⋅ Wojciech Kusa
→ What Makes an Encoding Good for Language Modeling?
Eugene Jang ⋅ Catherine Arnett ⋅ Terra Blevins
→ Dynamic Image Tokenization for Efficient VLMs
Yusen Peng ⋅ Tejas Naik ⋅ Valentin Hofmann ⋅ Yuki M Asano ⋅ Sachin Kumar
→ Are Non-BPE Tokenizers Ready to Improve LLM Numeracy?
Yizhan Huang ⋅ HUANG Nianchen ⋅ Yunxiang Li ⋅ Siu P Wong ⋅ Michael R Lyu
6:00 PM
Closing Remarks