paper-with-me

Papers

Knowledge Distillation of Black-Box Large Language Models

2024-01-13 · Hongzhan Chen, Ruijun Chen, Yuqi Yi, Xiaojun Quan, Chenliang Li, Ming Yan, Ji Zhang

Given the exceptional performance of proprietary large language models (LLMs) like GPT-4, recent research has increasingly focused on boosting the capabilities of smaller models through knowledge distillation (KD) from these powerful yet black-box teachers. While leveraging the high-quality outputs of these teachers is advantageous, the inaccessibility of their internal states often limits effective knowledge transfer. To overcome this limitation, we introduce Proxy-KD, a novel method that uses a proxy model to facilitate the efficient transfer of knowledge from black-box LLMs to smaller models. Our experiments show that Proxy-KD not only enhances the performance of KD from black-box teacher models but also surpasses traditional white-box KD techniques.~This approach presents a compelling new avenue for distilling knowledge from advanced LLMs.

📄 PDF Abstract BibTeX arXiv:2401.07013

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationTransfer Learning

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.

Similar Papers 제목 키워드 기반

Black-Box On-Policy Distillation of Large Language Models

2025-11-13 · Tianzhu Ye, Li Dong, Zewen Chi, Xun Wu 외 arxiv

Black-box distillation creates student large language models (LLMs) by learning from a proprietary teacher model's text outputs alone, without access to its internal logits or parameters. In this work, we introduce Gener…

Knowledge Distillation

From Images to Words: Efficient Cross-Modal Knowledge Distillation to Language Models from Black-box Teachers

2026-03-11 · Ayan Sengupta, Shantanu Dixit, Md Shad Akhtar, Tanmoy Chakraborty arxiv

Knowledge distillation (KD) methods are pivotal in compressing large pre-trained language models into smaller models, ensuring computational efficiency without significantly dropping performance. Traditional KD technique…

Natural Language UnderstandingComputational EfficiencyKnowledge Distillation

Intermediate Distillation: Data-Efficient Distillation from Black-Box LLMs for Information Retrieval

2024-06-18 · Zizhong Li, Haopeng Zhang, Jiawei Zhang

Recent research has explored distilling knowledge from large language models (LLMs) to optimize retriever models, especially within the retrieval-augmented generation (RAG) framework. However, most existing training meth…

Information RetrievalKnowledge DistillationQuestion AnsweringRAG+2

DDK: Distilling Domain Knowledge for Efficient Large Language Models

2024-07-23 · Jiaheng Liu, Chenchen Zhang, Jinyang Guo, Yuanxing Zhang 외

Despite the advanced intelligence abilities of large language models (LLMs) in various applications, they still face significant computational and storage demands. Knowledge Distillation (KD) has emerged as an effective …

Knowledge Distillation

SODA: Semi On-Policy Black-Box Distillation for Large Language Models

2026-04-04 · Xiwen Chen, Jingjing Wang, Wenhui Zhu, Peijie Qiu 외 arxiv

Black-box knowledge distillation for large language models presents a strict trade-off. Simple off-policy methods (e.g., sequence-level knowledge distillation) struggle to correct the student's inherent errors. Fully on-…

Knowledge Distillation