Challenge: Existing knowledge distillation methods require access to internal information of teachers . however, such information is not always accessible for large pre-trained language models .
Approach: They propose a method to estimate logits from the decision distributions using logits theoretically and empirically.
Outcome: The proposed method outperforms baselines on natural language understanding and machine reading comprehension datasets.

Similar Papers

Towards Understanding and Improving Knowledge Distillation for Neural Machine Translation (2023.acl-long)

Copied to clipboard

Challenge: Existing knowledge distillation techniques for neural machine translation lack special treatment on the top-1 information, which is limiting the potential of KD.
Approach: They propose a method to distill knowledge from top-1 predictions of teachers and a technique to infuse more additional knowledge by distilling on the data without ground-truth targets.
Outcome: The proposed method outperforms the vanilla word-level KD and outperfies the existing methods on three different students with different capacity gaps.
Pre-training Distillation for Large Language Models: A Design Space Exploration (2025.acl-long)

Copied to clipboard

Challenge: Knowledge distillation (KD) aims to transfer knowledge from a large teacher model to a smaller student model for model compression.
Approach: They extend knowledge distillation to the pre-training phase of large language models . they first conduct an experiment using a teacher LLM to distill a 1.9B student LLM .
Outcome: The proposed model can be used to distill a 1.9B student model using a teacher LLM.
Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs (2025.acl-long)

Copied to clipboard

Challenge: Knowledge distillation is a cost-effective technique to distill knowledge in Large Language Models, if the teacher output logits can be pre-computed and cached.
Approach: They propose an importance-sampling-based method which provides unbiased estimates, preserves the gradient in expectation, and requires storing significantly sparser logits.
Outcome: The proposed method enables faster training of student models with marginal overhead (10%) compared to cross-entropy based training, while maintaining competitive performance compared with full distillation.
Enhancing Knowledge Distillation of Large Language Models through Efficient Multi-Modal Distribution Alignment (2025.coling-main)

Copied to clipboard

Challenge: Existing knowledge distillation techniques for large language models are causing difficulties for student models to learn multi-modal probability distributions.
Approach: They propose a ranking loss-based knowledge distillation method that encourages consistency of the ranking of peak predictions between teacher and student models.
Outcome: The proposed method improves student models' ability to learn multi-modal distributions.
Knowledge Distillation for Language Models (2025.naacl-tutorial)

Copied to clipboard

Challenge: Knowledge distillation (KD) aims to transfer knowledge from a teacher to a student . this tutorial will cover topics ranging from LLM sequence compression to LLM self-distillation .
Approach: They propose to introduce intermediate-layer matching and prediction matching . they will then present advanced techniques such as reinforcement learning-based KD and multi-teacher distillation .
Outcome: This tutorial aims to provide participants with a comprehensive understanding of the techniques and applications of knowledge distillation for language models.
ReAugKD: Retrieval-Augmented Knowledge Distillation For Pre-trained Language Models (2023.acl-short)

Copied to clipboard

Challenge: Knowledge distillation (KD) is an effective compression technique to derive a smaller student model from a larger teacher model by transferring the knowledge embedded in the teacher's network.
Approach: They propose a framework and loss function that preserves the semantic similarities of teacher and student training examples to enable the student to retrieve from the knowledge base effectively.
Outcome: The proposed framework preserves the semantic similarities of teacher and student training examples to achieve state-of-the-art performance on the GLUE benchmark.
On the Generalization vs Fidelity Paradox in Knowledge Distillation (2025.findings-acl)

Copied to clipboard

Challenge: Knowledge distillation (KD) is a key technique for compressing large language models into smaller ones while preserving performance.
Approach: They propose to use knowledge distillation to compress large language models into smaller ones while preserving performance.
Outcome: The proposed technique improves the performance of smaller models by 10% while providing only marginal benefits for larger models.
Beyond Logits: Aligning Feature Dynamics for Effective Knowledge Distillation (2025.acl-long)

Copied to clipboard

Challenge: Knowledge distillation (KD) compresses large language models into lightweight versions called student models.
Approach: They propose to align the entire feature dynamics between teacher and student models by using two additional loss terms to achieve this.
Outcome: The proposed method matches the entire feature dynamics between teacher and student models rather than just the final states.
Tutoring Helps Students Learn Better: Improving Knowledge Distillation for BERT with Tutor Network (2022.emnlp-main)

Copied to clipboard

Challenge: Existing knowledge distillation approaches for language models have overlooked the difficulty of training examples.
Approach: They propose a framework that controls difficulty of training examples during pre-training by a tutor network.
Outcome: The proposed framework outperforms state-of-the-art KD methods with student models on the GLUE benchmark.
BiLD: Bi-directional Logits Difference Loss for Large Language Model Distillation (2025.coling-main)

Copied to clipboard

Challenge: Knowledge distillation (KD) is a method for reducing model size while preserving performance.
Approach: They propose a method to distill large language models at the logit level by transferring knowledge from a large teacher model to a smaller student model.
Outcome: The proposed method outperforms supervised fine-tuning, vanilla KL loss and five other distillation methods on 13 datasets.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations