paper-with-me

홈 › Papers

Compressing Sequences in the Latent Embedding Space: $K$-Token Merging for Large Language Models

2026-04-16 · Zihao Xu, John Harvill, Ziwei Fan, Yizhou Sun, Hao Ding, Hao Wang arxiv

Large Language Models (LLMs) incur significant computational and memory costs when processing long prompts, as full self-attention scales quadratically with input length. Token compression aims to address this challenge by reducing the number of tokens representing inputs. However, existing prompt-compression approaches primarily operate in token space and overlook inefficiencies in the latent embedding space. In this paper, we propose K-Token Merging, a latent-space compression framework that merges each contiguous block of K token embeddings into a single embedding via a lightweight encoder. The compressed sequence is processed by a LoRA-adapted LLM, while generation remains in the original vocabulary. Experiments on structural reasoning (Textualized Tree), sentiment classification (Amazon Reviews), and code editing (CommitPackFT) show that K-Token Merging lies on the Pareto frontier of performance vs. compression, achieving up to 75% input length reduction with minimal performance degradation. Code is available at https://github.com/shsjxzh/K-Token-Merging.

📄 PDF Abstract BibTeX arXiv:2604.15153

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

DWA-KD: Dual-Space Weighting and Time-Warped Alignment for Cross-Tokenizer Knowledge Distillation

2026-02-25 · Duc Trung Vu, Pham Khanh Chi, Dat Phi Van, Linh Ngo Van 외 arxiv

Knowledge Distillation (KD) has emerged as a crucial technique for compressing Large Language Models (LLMs). Although existing cross-tokenizer KD methods have made notable progress, their effectiveness remains constraine…

Knowledge Distillation

ProtFlow: Fast Protein Sequence Design via Flow Matching on Compressed Protein Language Model Embeddings

2025-04-15 · Zitai Kong, Yiheng Zhu, Yinlong Xu, Hanjing Zhou 외

The design of protein sequences with desired functionalities is a fundamental task in protein engineering. Deep generative methods, such as autoregressive models and diffusion models, have greatly accelerated the discove…

Language ModelingLanguage ModellingProtein DesignProtein Language Model

TLDR: Compressing Audio Tokens for Efficient Autoregressive Text-to-Speech

2026-06-08 · Yejin Lee, Junwon Moon, Hyoeun Kim, Hyunjin Choi 외 arxiv

Codec-based autoregressive (AR) speech language models have achieved strong text-to-speech (TTS) quality by modeling speech as sequences of discrete audio tokens with large pretrained backbones. However, this token-level…

Music2Latent2: Audio Compression with Summary Embeddings and Autoregressive Decoding

2025-01-29 · Marco Pasini, Stefan Lattner, George Fazekas

Efficiently compressing high-dimensional audio signals into a compact and informative latent space is crucial for various tasks, including generative modeling and music information retrieval (MIR). Existing audio autoenc…

Audio CompressionDenoisingInformation RetrievalMusic Information Retrieval+1

Do Llamas Work in English? On the Latent Language of Multilingual Transformers

2024-02-16 · Chris Wendler, Veniamin Veselovsky, Giovanni Monea, Robert West

We ask whether multilingual language models trained on unbalanced, English-dominated corpora use English as an internal pivot language -- a question of key importance for understanding how language models function and th…