{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/split-computing-and-early-exiting-for-deep","title":"Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges","arxiv_id":"2103.04505","date":"2021-03-08","proceeding":null,"authors":["Yoshitomo Matsubara","Marco Levorato","Francesco Restuccia"],"abstract":"Mobile devices such as smartphones and autonomous vehicles increasingly rely on deep neural networks (DNNs) to execute complex inference tasks such as image classification and speech recognition, among others. However, continuously executing the entire DNN on mobile devices can quickly deplete their battery. Although task offloading to cloud/edge servers may decrease the mobile device's computational burden, erratic patterns in channel quality, network, and edge server load can lead to a significant delay in task execution. Recently, approaches based on split computing (SC) have been proposed, where the DNN is split into a head and a tail model, executed respectively on the mobile device and on the edge server. Ultimately, this may reduce bandwidth usage as well as energy consumption. Another approach, called early exiting (EE), trains models to embed multiple \"exits\" earlier in the architecture, each providing increasingly higher target accuracy. Therefore, the trade-off between accuracy and delay can be tuned according to the current conditions or application demands. In this paper, we provide a comprehensive survey of the state of the art in SC and EE strategies by presenting a comparison of the most relevant approaches. We conclude the paper by providing a set of compelling research challenges.","url_abs":"https://arxiv.org/abs/2103.04505v4","url_pdf":"https://arxiv.org/pdf/2103.04505v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/JetRunner/PABEE","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/RyanXingQL/RBQE","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/alexa/wqa-cascade-transformers","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/autoliuweijie/FastBERT","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/castorini/deebert","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/castorini/earlyexiting-monobert","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/gmum/Zero-Time-Waste","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/jaryp/confidencebranchnetwok","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/jsiloto/adaptive-cod","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/kalviny/IMTA","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/kunglab/ddnn","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/mary-phuong/multiexit-distillation","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yangle15/RANet-pytorch","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yoshitomo-matsubara/head-network-distillation","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yoshitomo-matsubara/hnd-ghnd-object-detectors","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yoshitomo-matsubara/sc2-benchmark","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yoshitomo-matsubara/supervised-compression","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://gitlab.com/kunglab/branchynet","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"split-computing-and-early-exiting-for-deep","repo_url":"https://github.com/yoshitomo-matsubara/bottlefit-split_computing","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"autonomous-vehicles","task_name":"Autonomous Vehicles"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"early-exiting","method_name":"Early exiting"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2103.04505","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.04505"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/gmum/Zero-Time-Waste","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jsiloto/adaptive-cod","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kalviny/IMTA","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yoshitomo-matsubara/bottlefit-split_computing","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yoshitomo-matsubara/supervised-compression","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yoshitomo-matsubara/hnd-ghnd-object-detectors","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/castorini/deebert","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mary-phuong/multiexit-distillation","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://gitlab.com/kunglab/branchynet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yoshitomo-matsubara/sc2-benchmark","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/RyanXingQL/RBQE","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jaryp/confidencebranchnetwok","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/castorini/earlyexiting-monobert","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yoshitomo-matsubara/head-network-distillation","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kunglab/ddnn","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/autoliuweijie/FastBERT","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yangle15/RANet-pytorch","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/alexa/wqa-cascade-transformers","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/JetRunner/PABEE","reach":{"status":"unanswered"}}],"summary":{"ran_draft_wrong":2},"by_repo_kind":{"named_in_paper":{"samples":2,"ran":2,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"87cde55ede2fa968","entry":"entropy","repo":"castorini/earlyexiting-monobert","repo_kind":"named_in_paper","path":"transformers/modeling_highway_bert.py","file_url":"https://github.com/castorini/earlyexiting-monobert/blob/HEAD/transformers/modeling_highway_bert.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"87cde55ede2fa968"}},{"code_sha256_prefix":"cc16728fc65f6915","entry":"normal_shannon_entropy","repo":"autoliuweijie/FastBERT","repo_kind":"named_in_paper","path":"run_fastbert.py","file_url":"https://github.com/autoliuweijie/FastBERT/blob/HEAD/run_fastbert.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cc16728fc65f6915"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}