ICML 2025 Workshop TokShop Submissions

Byte Latent Transformer: Patches Scale Better Than Tokens
ICML 2025 Workshop TokShop Submission19 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
BPE Stays on SCRIPT: Structured Encoding for Robust Multilingual Pretokenization
ICML 2025 Workshop TokShop Submission16 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Conditional Unigram Tokenization with Parallel Data
ICML 2025 Workshop TokShop Submission15 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Overcoming Vocabulary Constraints with Pixel-level Fallback
ICML 2025 Workshop TokShop Submission14 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
zip2zip: Inference-Time Adaptive Vocabularies for Language Models via Token Compression
ICML 2025 Workshop TokShop Submission13 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Tokenisation is NP-Complete
ICML 2025 Workshop TokShop Submission12 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Watermarking Autoregressive Image Generation
ICML 2025 Workshop TokShop Submission11 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Robust Noise Attenuation via Adaptive Pooling of Transformer Outputs
ICML 2025 Workshop TokShop Submission10 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression
ICML 2025 Workshop TokShop Submission9 Authors
- Published: 10 Jun 2025, Last Modified: 13 Jun 2025
- TokShop
- Readers: Everyone
Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8
ICML 2025 Workshop TokShop Submission8 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
Is Your LLM Overcharging You? Tokenization, Transparency, and Incentives
ICML 2025 Workshop TokShop Submission7 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
CAT: Content-Adaptive Image Tokenization
ICML 2025 Workshop TokShop Submission5 Authors
- Published: 10 Jun 2025, Last Modified: 11 Jun 2025
- TokShop
- Readers: Everyone
QuickMerge++: Token Merging with Autoregressive Prior
ICML 2025 Workshop TokShop Submission4 Authors
- Published: 10 Jun 2025, Last Modified: 12 Jun 2025
- TokShop
- Readers: Everyone