Papers › CascadeBERT: Accelerating Inference of Pre-trained Language Models via Calibrated...

CascadeBERT: Accelerating Inference of Pre-trained Language Models via Calibrated Complete Models Cascade

29 Dec 2020Findings (EMNLP) 2021 11arXiv:2012.14682archive 2025-07-28

Lei LI, Yankai Lin, Deli Chen, Shuhuai Ren, Peng Li, Jie zhou, Xu sun

Dynamic early exiting aims to accelerate the inference of pre-trained language models (PLMs) by emitting predictions in internal layers without passing through the entire model. In this paper, we empirically analyze the working mechanism of dynamic early exiting and find that it faces a performance bottleneck under high speed-up ratios. On one hand, the PLMs' representations in shallow layers lack high-level semantic information and thus are not sufficient for accurate predictions. On the other hand, the exiting decisions made by internal classifiers are unreliable, leading to wrongly emitted early predictions. We instead propose a new framework for accelerating the inference of PLMs, CascadeBERT, which dynamically selects proper-sized and complete models in a cascading manner, providing comprehensive representations for predictions. We further devise a difficulty-aware objective, encouraging the model to output the class probability that reflects the real difficulty of each instance for a more reliable cascading mechanism. Experimental results show that CascadeBERT can achieve an overall 15\% improvement under 4× speed-up compared with existing dynamic early exiting methods on six classification tasks, yielding more calibrated and accurate predictions.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

lancopku/cascadebert officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Knowledge DistillationModel Selection

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Early exiting

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections