paper-with-me

Papers

Task-agnostic Distillation of Encoder-Decoder Language Models

2023-05-21 · Chen Zhang, Yang Yang, Jingang Wang, Dawei Song

Finetuning pretrained language models (LMs) have enabled appealing performance on a diverse array of tasks. The intriguing task-agnostic property has driven a shifted focus from task-specific to task-agnostic distillation of LMs. While task-agnostic, compute-efficient, performance-preserved LMs can be yielded by task-agnostic distillation, previous studies mainly sit in distillation of either encoder-only LMs (e.g., BERT) or decoder-only ones (e.g., GPT) yet largely neglect that distillation of encoder-decoder LMs (e.g., T5) can posit very distinguished behaviors. Frustratingly, we discover that existing task-agnostic distillation methods can fail to handle the distillation of encoder-decoder LMs. To the demand, we explore a few paths and uncover a path named as MiniEnD that successfully tackles the distillation of encoder-decoder LMs in a task-agnostic fashion. We examine MiniEnD on language understanding and abstractive summarization. The results showcase that MiniEnD is generally effective and is competitive compared to other alternatives. We further scale MiniEnD up to distillation of 3B encoder-decoder language models with interpolated distillation. The results imply the opportunities and challenges in distilling large language models (e.g., LLaMA).

📄 PDF Abstract BibTeX arXiv:2305.12330

Code (0)

등록된 구현이 없습니다.

Tasks

Abstractive Text SummarizationDecoder

Methods 이 논문이 사용한 방법론

fail 설명 없음

Similar Papers 제목 키워드 기반

Generic-to-Specific Distillation of Masked Autoencoders

2023-02-28 · CVPR 2023 1 · Wei Huang, Zhiliang Peng, Li Dong, Furu Wei 외

Large vision Transformers (ViTs) driven by self-supervised pre-training mechanisms achieved unprecedented progress. Lightweight ViT models limited by the model capacity, however, benefit little from those pre-training me…

Decoderimage-classificationImage ClassificationKnowledge Distillation+3

Dynamic Self-Distillation via Previous Mini-batches for Fine-tuning Small Language Models

2024-11-25 · Yao Fu, Yin Yu, Xiaotian Han, Runchao Li 외

Knowledge distillation (KD) has become a widely adopted approach for compressing large language models (LLMs) to reduce computational costs and memory footprints. However, the availability of complex teacher models is a …

Knowledge DistillationNatural Language UnderstandingText Generation

EdgeSAM: Prompt-In-the-Loop Distillation for On-Device Deployment of SAM

2023-12-11 · Chong Zhou, Xiangtai Li, Chen Change Loy, Bo Dai

This paper presents EdgeSAM, an accelerated variant of the Segment Anything Model (SAM), optimized for efficient execution on edge devices with minimal compromise in performance. Our approach involves distilling the orig…

Decoder

Decoder-based Sense Knowledge Distillation

2026-02-25 · Qitong Wang, Mohammed J. Zaki, Georgios Kollias, Vasileios Kalantzis arxiv

Large language models (LLMs) learn contextual embeddings that capture rich semantic information, yet they often overlook structured lexical knowledge such as word senses and relationships. Prior work has shown that incor…

Knowledge Distillation

Return of the Encoder: Maximizing Parameter Efficiency for SLMs

2025-01-27 · Mohamed Elfeki, Rui Liu, Chad Voegele

The dominance of large decoder-only language models has overshadowed encoder-decoder architectures, despite their fundamental efficiency advantages in sequence processing. For small language models (SLMs) - those with 1 …

Computational EfficiencyCPUDecoderGPU+1