arXiv Computation and Language October 12, 2026 By Saibo Geng, Andrea Gentilini, Xinxian Ma, Nathan Ranchin, Yunzhen Yao, Robert West

zip2zip++: Towards Inference-Time Token Compression with Minimal Quality Loss

Read the original on arXiv Computation and Language →

The paper introduces zip2zip++, a method for inference‑time token compression that reduces token counts using a dynamic Lempel‑Ziv‑Welch (LZW) codec while preserving language model quality. It identifies four problems with prior LZW‑based compression—embedding conflict, improper RoPE application, harmful digit merges, and segmentation ambiguity—and proposes solutions: separate input/output embeddings, size‑aware RoPE, and a selective‑merge LZW codec. Experiments on four models show that these changes halve the quality gap in perplexity and reasoning accuracy compared to the uncompressed baseline, achieving similar compression ratios.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI

Aug 13

Diffuse to Compress: Leveraging Diffusion LMs for Lossless Compression

arXiv:2608. 11249v1 Announce Type: cross Abstract: We study the problem of lossless text compression, motivated by the rapid growth in the collection and storage of digital textual data - including plain text, source code, and structured formats such as XML - and by recent advances in neural language model-based compression.

By Angelo Nardone, Paolo Ferragina

llms diffusion efficiency benchmarks

arXiv AI

Sep 24

Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models

arXiv:2609.27510v1 Announce Type: cross Abstract: Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and...

By Kaifeng Tan, Yudong Li, Linlin Shen

llms benchmarks

arXiv Computation and Language

Sep 15

Lexical Prompt Compression for Large Language Models: A Training-Free, Deterministic Pipeline with Empirical Pareto Analysis Across Eleven Task Categories

arXiv:2609.13154v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have made prompts increasingly large and complex. Techniques such as chain-of-thought reasoning (Wei et...

By Shamin Chokshi

llms nlp efficiency benchmarks

arXiv AI

Jul 31

ARC-Encoder: learning compressed text representations for large language models

arXiv:2510. 20535v2 Announce Type: replace-cross Abstract: Recent techniques such as retrieval-augmented generation or chain-of-thought reasoning have led to longer contexts and increased inference costs.

By Hippolyte Pilchen, Edouard Grave, Patrick P\'erez

llms rag fine-tuning benchmarks

Hugging Face Trending Papers

Sep 23

Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models

Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and undermining the reliability of evaluation results....

llms benchmarks

arXiv AI

Sep 28

Achieving Tokenizer Flexibility in Language Models through Heuristic Adaptation and Supertoken Learning

The paper introduces TokenAdapt, a model‑agnostic tokenizer transplantation method that uses a hybrid heuristic to initialize new token embeddings, and a novel pre‑tokenization learning approach for multi‑word Supertokens to improve compression. TokenAdapt combines local subword decomposition and global semantic similarity to preserve semantics while reducing retraining needs. Empirical results show that TokenAdapt outperforms existing baselines such as Transtokenizer and ReTok, achieving lower perplexity ratios and significant compression gains.

By Shaurya Sharthak, Vinayak Pahalwan, Adithya Kamath, Adarsh Shirawalmath

llms rag nlp fine-tuning

One email a morning, machine-written

Related stories

Diffuse to Compress: Leveraging Diffusion LMs for Lossless Compression

Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models

Lexical Prompt Compression for Large Language Models: A Training-Free, Deterministic Pipeline with Empirical Pareto Analysis Across Eleven Task Categories

ARC-Encoder: learning compressed text representations for large language models

Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models

Achieving Tokenizer Flexibility in Language Models through Heuristic Adaptation and Supertoken Learning