paper-with-me

홈 › Papers

Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing

2026-03-18 · Raghavv Goel, Mukul Gagrani, Mingu Lee, Chris Lott arxiv

Large Language Models (LLMs) possess latent multi-token prediction (MTP) abilities despite being trained only for next-token generation. We introduce ESP (Embedding-Space Probing), a simple and training-free MTP method that probes an LLM using on-the-fly mask tokens drawn from its embedding space, enabling parallel future-token prediction without modifying weights or relying on draft models. ESP constructs a speculative token tree by sampling Top-K candidates from mask-token logits and applies a lightweight pruning rule to retain high-probability continuations. During generation, predictions are verified in parallel, yielding lossless decoding while significantly reducing model calls and increasing token throughput. ESP consistently outperforms existing training-free baselines, improving acceptance length by 7-11% over LADE on LLaMA3 and 7-8% on Qwen3, and increasing throughput by up to 15-19% over the strongest baseline. Finally, we provide theoretical insight and empirical evidence showing that decoder layers naturally align mask-token representations with next-token states, enabling accurate multi-step prediction without retraining or auxiliary models.

📄 PDF Abstract BibTeX arXiv:2603.17942

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs

2026-01-03 · Yixuan Tang, Yi Yang arxiv

While LLMs are powerful embedding backbones, their application in training-free settings faces two structural challenges: causal attention restricts early tokens from accessing subsequent context, and the next-token pred…

Representation Learning

LLM-based Embeddings: Attention Values Encode Sentence Semantics Better Than Hidden States

2026-02-02 · Yeqin Zhang, Yunfei Wang, Jiaxuan Chen, Ke Qin 외 arxiv

Sentence representations are foundational to many Natural Language Processing (NLP) applications. While recent methods leverage Large Language Models (LLMs) to derive sentence representations, most rely on final-layer hi…

ArcMark: Distortion-Free Multi-Byte LLM Watermark via Optimal Transport

2026-02-06 · Atefeh Gilani, Sajani Vithana, Carol Xuan Long, Oliver Kosut 외 arxiv

Watermarking is an important tool for promoting the responsible use of large language models (LLMs). Existing watermarks insert a signal into generated tokens that either flags LLM-generated text (zero-bit watermarking) …

Context-level Language Modeling by Learning Predictive Context Embeddings

2025-10-23 · Beiya Dai, Yuliang Liu, Daozheng Xue, Yunchong Song 외 arxiv

We propose ContextLM, a framework that implicitly learns multi-token prediction by augmenting standard pretraining with an intrinsic next-context prediction objective. ContextLM builds a language model on top of context …

SeLaR: Selective Latent Reasoning in Large Language Models

2026-04-09 · Renyu Fu, Guibo Luo arxiv

Chain-of-Thought (CoT) has become a cornerstone of reasoning in large language models, yet its effectiveness is constrained by the limited expressiveness of discrete token sampling. Recent latent reasoning approaches att…