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
In Syntology name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.
Near matches the lookup lists, each a different name: A Simple Early Exiting Framework; Performance Control Early Exiting (PCEE); Hash-based Early Exiting approach (HASHEE); Fast and Robust Early-Exiting (FREE) framework; Learning to Weight Samples for Dynamic Early-exiting Networks; early dropout; Early Cutting; Early-Bird Decoding.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 66 · with a code link: 28 · where Syntology ran a sample: 9 (7 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (9 of 66 tagged: 7 with a run with no instrument failure, 2 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 66 of 66, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
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 official: harvested, nothing ran · 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 2 honoured, 2 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 5 harvested samples) · 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 official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified (of 6 harvested samples) · 6 pointer-only (licence)
-
LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding 25 Apr 2024 · 1 repository · arXiv:2404.16710Syntology official (archive's flag): 4 ran · 4 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified (of 5 harvested samples) · 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 official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified (of 5 harvested samples) · 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
-
Sustainable Edge Intelligence Through Energy-Aware Early Exiting 23 May 2023 · 0 repositories · arXiv:2305.14094
-
F-PABEE: Flexible-patience-based Early Exiting for Single-label and Multi-label text Classification Tasks 21 May 2023 · 0 repositories · arXiv:2305.11916
-
SmartBERT: A Promotion of Dynamic Early Exiting Mechanism for Accelerating BERT Inference 16 Mar 2023 · 0 repositories · arXiv:2303.09266Syntology 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; every one of the 3 samples that ran constructed an object rather than computing a result (of 4 harvested samples)
-
Adaptive Deep Neural Network Inference Optimization with EENet 15 Jan 2023 · 1 repository · arXiv:2301.07099
-
AdaEnsemble: Learning Adaptively Sparse Structured Ensemble Network for Click-Through Rate Prediction 6 Jan 2023 · 0 repositories · arXiv:2301.08353
-
HADAS: Hardware-Aware Dynamic Neural Architecture Search for Edge Performance Scaling 6 Dec 2022 · 1 repository · arXiv:2212.03354
-
You Need Multiple Exiting: Dynamic Early Exiting for Accelerating Unified Vision Language Model 21 Nov 2022 · 1 repository · arXiv:2211.11152
-
COST-EFF: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models 27 Oct 2022 · 1 repository · arXiv:2210.15523
-
Unsupervised Early Exit in DNNs with Multiple Exits 20 Sep 2022 · 1 repository · arXiv:2209.09480
-
Improving the Performance of DNN-based Software Services using Automated Layer Caching 18 Sep 2022 · 0 repositories · arXiv:2209.08625
-
Confident Adaptive Language Modeling 14 Jul 2022 · 0 repositories · arXiv:2207.07061
-
A Simple Hash-Based Early Exiting Approach For Language Understanding and Generation 3 Mar 2022 · 1 repository · arXiv:2203.01670Syntology official (archive's flag): 6 ran · 6 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 7 harvested samples)
-
Magic Pyramid: Accelerating Inference with Early Exiting and Token Pruning 16 Jan 2022 · 0 repositories
-
PCEE-BERT: Accelerating BERT Inference via Patient and Confident Early Exiting 16 Jan 2022 · 0 repositories
-
VEE-BERT: Accelerating BERT Inference for Named Entity Recognition via Vote Early Exiting 16 Jan 2022 · 0 repositories
-
DeeCap: Dynamic Early Exiting for Efficient Image Captioning 1 Jan 2022 · 1 repository
-
A Simple Hash-Based Early Exiting Approach For Language Understanding and Generation 16 Nov 2021 · 0 repositories
-
Magic Pyramid: Accelerating Inference with Early Exiting and Token Pruning 30 Oct 2021 · 0 repositories · arXiv:2111.00230
-
Wacky Weights in Learned Sparse Representations and the Revenge of Score-at-a-Time Query Evaluation 22 Oct 2021 · 0 repositories · arXiv:2110.11540
-
Towards Efficient NLP: A Standard Evaluation and A Strong Baseline 13 Oct 2021 · 1 repository · arXiv:2110.07038
-
Accelerating BERT Inference for Sequence Labeling via Early-Exit 28 May 2021 · 1 repository · arXiv:2105.13878
-
Early Exiting with Ensemble Internal Classifiers 28 May 2021 · 0 repositories · arXiv:2105.13792
-
Single-Layer Vision Transformers for More Accurate Early Exits with Less Overhead 19 May 2021 · 0 repositories · arXiv:2105.09121
-
FrameExit: Conditional Early Exiting for Efficient Video Recognition 27 Apr 2021 · 1 repository · arXiv:2104.13400Syntology official (archive's flag): 6 ran · 6 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 4 where Syntology's instrument failed) · 3 unverified (of 9 harvested samples) · 9 pointer-only (licence)
-
BERxiT: Early Exiting for BERT with Better Fine-Tuning and Extension to Regression 1 Apr 2021 · 1 repository
-
Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges 8 Mar 2021 · 19 repositories · arXiv:2103.04505Syntology 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified (of 2 harvested samples) · 2 pointer-only (licence)
-
RomeBERT: Robust Training of Multi-Exit BERT 24 Jan 2021 · 1 repository · arXiv:2101.09755
-
CascadeBERT: Accelerating Inference of Pre-trained Language Models via Calibrated Complete Models Cascade 29 Dec 2020 · 1 repository · arXiv:2012.14682
-
HAPI: Hardware-Aware Progressive Inference 10 Aug 2020 · 0 repositories · arXiv:2008.03997
-
BERT Loses Patience: Fast and Robust Inference with Early Exit 7 Jun 2020 · 1 repository · arXiv:2006.04152
-
Query-level Early Exit for Additive Learning-to-Rank Ensembles 30 Apr 2020 · 0 repositories · arXiv:2004.14641
-
DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference 27 Apr 2020 · 3 repositories · arXiv:2004.12993Syntology official (archive's flag): 1 ran · 5 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 6 harvested samples) · 5 pointer-only (licence)
-
Fast Intent Classification for Spoken Language Understanding 3 Dec 2019 · 1 repository · arXiv:1912.01728
-
Depth-Adaptive Transformer 22 Oct 2019 · 0 repositories · arXiv:1910.10073
-
Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge Computing 4 Oct 2019 · 0 repositories · arXiv:1910.05316
-
BranchyNet: Fast Inference via Early Exiting from Deep Neural Networks 6 Sep 2017 · 3 repositories · arXiv:1709.01686
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