REVIEW 6 cited by
Rethinking Kullback-Leibler Divergence in Knowledge Distillation for Large Language Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Kullback-Leiber divergence has been widely used in Knowledge Distillation (KD) to compress Large Language Models (LLMs). Contrary to prior assertions that reverse Kullback-Leibler (RKL) divergence is mode-seeking and thus preferable over the mean-seeking forward Kullback-Leibler (FKL) divergence, this study empirically and theoretically demonstrates that neither mode-seeking nor mean-seeking properties manifest in KD for LLMs. Instead, RKL and FKL are found to share the same optimization objective and both converge after a sufficient number of epochs. However, due to practical constraints, LLMs are seldom trained for such an extensive number of epochs. Meanwhile, we further find that RKL focuses on the tail part of the distributions, while FKL focuses on the head part at the beginning epochs. Consequently, we propose a simple yet effective Adaptive Kullback-Leiber (AKL) divergence method, which adaptively allocates weights to combine FKL and RKL. Metric-based and GPT-4-based evaluations demonstrate that the proposed AKL outperforms the baselines across various tasks and improves the diversity and quality of generated responses. Codes are available at \href{https://github.com/wutaiqiang/LLM_KD_AKL}{github}.
Forward citations
Cited by 6 Pith papers
-
Weak-to-Strong Generalization via Direct On-Policy Distillation
Transferring the log-ratio of a small model's pre-RL and post-RL checkpoints provides a dense implicit reward that improves stronger student models at a fraction of the cost of direct RL.
-
GenRecal: Generation after Recalibration from Large to Small Vision-Language Models
A learnable Recalibrator bridges different tokenizers so that small VLMs can distill knowledge from any large VLM, improving their benchmark scores.
-
TAID: Temporally Adaptive Interpolated Distillation for Efficient Knowledge Transfer in Language Models
TAID improves knowledge distillation for language models by gradually shifting the student's training target from its own distribution to the teacher's distribution over time.
-
Self-Evolution Knowledge Distillation for LLM-based Machine Translation
A token-adaptive distillation method that mixes teacher and ground-truth targets only for hard tokens yields consistent BLEU gains in LLM translation.
-
Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models
A multi-level optimal transport loss combining sequence-level ranking, top-k truncation, and Sinkhorn sequence distance outperforms earlier cross-tokenizer distillation losses on QA and summarization.
-
SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought
SCOUT combines progressive distillation with a cross-attention module to make recursive latent reasoning work through fine-tuning, yielding up to 1.8% accuracy gains over standard fine-tuning.
Discussion (0). Continue with ORCID to comment.