paper-with-me

Papers

BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models

2024-01-23 · Feng Lin, Hanling Yi, Hongbin Li, Yifan Yang, Xiaotian Yu, Guangming Lu, Rong Xiao

Large language models (LLMs) commonly employ autoregressive generation during inference, leading to high memory bandwidth demand and consequently extended latency. To mitigate this inefficiency, we present Bi-directional Tuning for lossless Acceleration (BiTA), an innovative method expediting LLMs via streamlined semi-autoregressive generation and draft verification. Inspired by the concept of prompt tuning, we enhance LLMs with a parameter-efficient design called bi-directional tuning for the capability in semi-autoregressive generation. Employing efficient tree-based decoding, the models perform draft candidate generation and verification in parallel, ensuring outputs identical to their autoregressive counterparts under greedy sampling. BiTA serves as a lightweight plug-in module, seamlessly boosting the inference efficiency of existing LLMs without requiring additional assistance models or incurring significant extra memory costs. Applying the proposed BiTA, LLaMA-2-70B-Chat achieves a 2.7$\times$ speedup on the MT-Bench benchmark. Extensive experiments confirm our method surpasses state-of-the-art acceleration techniques.

📄 PDF Abstract BibTeX arXiv:2401.12522

Code (1)

linfeng93/bita 공식 구현 pytorch

Similar Papers 제목 키워드 기반

InfoBatch: Lossless Training Speed Up by Unbiased Dynamic Data Pruning

2023-03-08 · Ziheng Qin, Kai Wang, Zangwei Zheng, Jianyang Gu 외

Data pruning aims to obtain lossless performances with less overall cost. A common approach is to filter out samples that make less contribution to the training. This could lead to gradient expectation bias compared to t…

Semantic Segmentation

Lossless Acceleration of Large Language Models with Hierarchical Drafting based on Temporal Locality in Speculative Decoding

2025-02-08 · Sukmin Cho, Sangjin Choi, Taeho Hwang, Jeongyeon Seo 외

Accelerating inference in Large Language Models (LLMs) is critical for real-time interactions, as they have been widely incorporated into real-world services. Speculative decoding, a fully algorithmic solution, has gaine…

Dynamic Video Generation: Shaping Video Generation Across Time and Space

2026-05-20 · Shikang Zheng, Jingkai Huang, Jiacheng Liu, Guantao Chen 외 arxiv

Diffusion models have achieved impressive performance in video generation, but their iterative denoising process remains computationally expensive due to the large number of tokens processed at each timestep. Recently, p…

Video Generation

SampleAttention: Near-Lossless Acceleration of Long Context LLM Inference with Adaptive Structured Sparse Attention

2024-06-17 · Qianchao Zhu, Jiangfei Duan, Chang Chen, Siran Liu 외

Large language models (LLMs) now support extremely long context windows, but the quadratic complexity of vanilla attention results in significantly long Time-to-First-Token (TTFT) latency. Existing approaches to address …

TIDE: Efficient and Lossless MoE Diffusion LLM Inference with I/O-aware Expert Offload

2026-05-19 · Zhiben Chen, Youpeng Zhao, Yang Sui, Jun Wang 외 arxiv

Diffusion Large Language Models (dLLMs) have emerged as a competitive alternative to autoregressive (AR) models, offering better hardware utilization and bidirectional context through parallel block-level decoding. Howev…