Methods › General › Loss Functions › Early exiting
Early exiting using confidence measures
Early exiting
Introduced by Surat Teerapittayanon et al. in BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Exit whenever the model is confident enough allowing early exiting from hidden layers
Papers archive 2025-07-28
30 shown of 66, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Early Stopping Tabular In-Context Learning 26 Jun 2025 · 0 repositories · arXiv:2506.21387
-
Languages in Multilingual Speech Foundation Models Align Both Phonetically and Semantically 26 May 2025 · 0 repositories · arXiv:2505.19606
-
System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts 25 May 2025 · 0 repositories · arXiv:2505.18962
-
Performance Control in Early Exiting to Deploy Large Models at the Same Cost of Smaller Ones 26 Dec 2024 · 0 repositories · arXiv:2412.19325
-
COSEE: Consistency-Oriented Signal-Based Early Exiting via Calibrated Sample Weighting Mechanism 17 Dec 2024 · 1 repository · arXiv:2412.13236
-
A Stitch in Time Saves Nine: Small VLM is a Precise Guidance for Accelerating Large VLMs 4 Dec 2024 · 1 repository · arXiv:2412.03324Syntology ran 4 of 5 samples · 1 unverified · 5 pointer-only (licence)
-
Energy-Aware Dynamic Neural Inference 4 Nov 2024 · 0 repositories · arXiv:2411.02471
-
BERTer: The Efficient One 19 Jul 2024 · 0 repositories · arXiv:2407.14039
-
Tracing Representation Progression: Analyzing and Enhancing Layer-Wise Similarity 20 Jun 2024 · 0 repositories · arXiv:2406.14479
-
Predicting Probabilities of Error to Combine Quantization and Early Exiting: QuEE 20 Jun 2024 · 0 repositories · arXiv:2406.14404
-
RAEE: A Robust Retrieval-Augmented Early Exiting Framework for Efficient Inference 24 May 2024 · 0 repositories · arXiv:2405.15198
-
A Comprehensive Survey of Accelerated Generation Techniques in Large Language Models 15 May 2024 · 1 repository · arXiv:2405.13019
-
Computational Efficient Width-Wise Early Exiting in Wireless Communication Systems 6 May 2024 · 0 repositories · arXiv:2405.03222
-
Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting 29 Apr 2024 · 1 repository · arXiv:2404.18911Syntology ran 4 of 6 samples · 2 unverified · 6 pointer-only (licence)
-
LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding 25 Apr 2024 · 1 repository · arXiv:2404.16710Syntology ran 4 of 5 samples · 1 unverified · 5 pointer-only (licence)
-
Efficient Transformer Encoders for Mask2Former-style models 23 Apr 2024 · 1 repository · arXiv:2404.15244
-
Tiny Models are the Computational Saver for Large Models 26 Mar 2024 · 1 repository · arXiv:2403.17726
-
Understanding the Training Speedup from Sampling with Approximate Losses 10 Feb 2024 · 0 repositories · arXiv:2402.07052
-
EERO: Early Exit with Reject Option for Efficient Classification with limited budget 6 Feb 2024 · 0 repositories · arXiv:2402.03779
-
DE³-BERT: Distance-Enhanced Early Exiting for BERT based on Prototypical Networks 3 Feb 2024 · 0 repositories · arXiv:2402.05948
-
ConsistentEE: A Consistent and Hardness-Guided Early Exiting Method for Accelerating Language Models Inference 19 Dec 2023 · 1 repository · arXiv:2312.11882
-
EE-LLM: Large-Scale Training and Inference of Early-Exit Large Language Models with 3D Parallelism 8 Dec 2023 · 1 repository · arXiv:2312.04916Syntology ran 3 of 5 samples · 2 unverified · 5 pointer-only (licence)
-
Adaptive Early Exiting for Collaborative Inference over Noisy Wireless Channels 29 Nov 2023 · 0 repositories · arXiv:2311.18098
-
Accelerating LLaMA Inference by Enabling Intermediate Layer Decoding via Instruction Tuning with LITE 28 Oct 2023 · 0 repositories · arXiv:2310.18581
-
AdaDiff: Accelerating Diffusion Models through Step-Wise Adaptive Computation 29 Sep 2023 · 0 repositories · arXiv:2309.17074
-
Dynamic Early Exiting Predictive Coding Neural Networks 5 Sep 2023 · 0 repositories · arXiv:2309.02022
-
Using Early Exits for Fast Inference in Automatic Modulation Classification 22 Aug 2023 · 0 repositories · arXiv:2308.11100
-
LGViT: Dynamic Early Exiting for Accelerating Vision Transformer 1 Aug 2023 · 1 repository · arXiv:2308.00255
-
SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inference 5 Jul 2023 · 0 repositories · arXiv:2307.02628
-
CSI-Based Efficient Self-Quarantine Monitoring System Using Branchy Convolution Neural Network 24 May 2023 · 0 repositories · arXiv:2306.01756
Tasks archive 2025-07-28
20 shown of 86 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
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