paper-with-me

홈 › Papers

Baby Llama: knowledge distillation from an ensemble of teachers trained on a small dataset with no performance penalty

2023-08-03 · Inar Timiryasov, Jean-Loup Tastet

We present our submission to the BabyLM challenge, whose goal was to improve the sample efficiency of language models. We trained an ensemble consisting of a GPT-2 and small LLaMA models on the developmentally-plausible, 10M-word BabyLM dataset, then distilled it into a small, 58M-parameter LLaMA model, which exceeds in performance both of its teachers as well as a similar model trained without distillation. This suggests that distillation can not only retain the full performance of the teacher model when the latter is trained on a sufficiently small dataset; it can exceed it, and lead to significantly better performance than direct training.

📄 PDF Abstract BibTeX arXiv:2308.02019

Code (1)

timinar/babyllama 공식 구현 pytorch

Tasks

Knowledge Distillation

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Adam 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…

Similar Papers 제목 키워드 기반

BabyLlama-2: Ensemble-Distilled Models Consistently Outperform Teachers With Limited Data

2024-09-25 · Jean-Loup Tastet, Inar Timiryasov

We present BabyLlama-2, a 345 million parameter model distillation-pretrained from two teachers on a 10 million word corpus for the BabyLM competition. On BLiMP and SuperGLUE benchmarks, BabyLlama-2 outperforms baselines…

Choosy Babies Need One Coach: Inducing Mode-Seeking Behavior in BabyLlama with Reverse KL Divergence

2024-10-29 · Shaozhen Shi, Yevgen Matusevych, Malvina Nissim

This study presents our submission to the Strict-Small Track of the 2nd BabyLM Challenge. We use a teacher-student distillation setup with the BabyLLaMa model (Timiryasov and Tastet, 2023) as a backbone. To make the stud…

Unified and Effective Ensemble Knowledge Distillation

2022-04-01 · Chuhan Wu, Fangzhao Wu, Tao Qi, Yongfeng Huang

Ensemble knowledge distillation can extract knowledge from multiple teacher models and encode it into a single student model. Many existing methods learn and distill the student model on labeled data only. However, the t…

Knowledge DistillationTransfer Learning

Adaptive Group Robust Ensemble Knowledge Distillation

2024-11-22 · Patrik Kenfack, Ulrich Aïvodji, Samira Ebrahimi Kahou

Neural networks can learn spurious correlations in the data, often leading to performance disparity for underrepresented subgroups. Studies have demonstrated that the disparity is amplified when knowledge is distilled fr…

Knowledge Distillation

Agree to Disagree: Adaptive Ensemble Knowledge Distillation in Gradient Space

2020-12-01 · NeurIPS 2020 12 · Shangchen Du, Shan You, Xiaojie Li, Jianlong Wu 외

Distilling knowledge from an ensemble of teacher models is expected to have a more promising performance than that from a single one. Current methods mainly adopt a vanilla average rule, i.e., to simply take the average …

DiversityKnowledge Distillation