{"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/neural-speed-reading-via-skim-rnn","title":"Neural Speed Reading via Skim-RNN","arxiv_id":"1711.02085","date":"2017-11-06","proceeding":"ICLR 2018 1","authors":["Minjoon Seo","Sewon Min","Ali Farhadi","Hannaneh Hajishirzi"],"abstract":"Inspired by the principles of speed reading, we introduce Skim-RNN, a\nrecurrent neural network (RNN) that dynamically decides to update only a small\nfraction of the hidden state for relatively unimportant input tokens. Skim-RNN\ngives computational advantage over an RNN that always updates the entire hidden\nstate. Skim-RNN uses the same input and output interfaces as a standard RNN and\ncan be easily used instead of RNNs in existing models. In our experiments, we\nshow that Skim-RNN can achieve significantly reduced computational cost without\nlosing accuracy compared to standard RNNs across five different natural\nlanguage tasks. In addition, we demonstrate that the trade-off between accuracy\nand speed of Skim-RNN can be dynamically controlled during inference time in a\nstable manner. Our analysis also shows that Skim-RNN running on a single CPU\noffers lower latency compared to standard RNNs on GPUs.","url_abs":"http://arxiv.org/abs/1711.02085v3","url_pdf":"http://arxiv.org/pdf/1711.02085v3.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":"neural-speed-reading-via-skim-rnn","repo_url":"https://github.com/ht1221/leap-lstm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":null,"task_name":"CPU"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.02085","atlas_url":"https://app.syntology.ai/?focus=1711.02085","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1711.02085"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/ht1221/leap-lstm","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":4},"by_repo_kind":{"listed":{"samples":4,"ran":0,"repositories":1}},"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":0,"samples":[{"code_sha256_prefix":"e7121edc84f532e9","entry":"gumbel_softmax_sample","repo":"ht1221/leap-lstm","repo_kind":"listed","path":"skiplstm.py","file_url":"https://github.com/ht1221/leap-lstm/blob/HEAD/skiplstm.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e7121edc84f532e9"}},{"code_sha256_prefix":"35b3d781fb8b197b","entry":"sample_gumbel","repo":"ht1221/leap-lstm","repo_kind":"listed","path":"skiplstm.py","file_url":"https://github.com/ht1221/leap-lstm/blob/HEAD/skiplstm.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"35b3d781fb8b197b"}},{"code_sha256_prefix":"287ebfa206000646","entry":"update_best_acc","repo":"ht1221/leap-lstm","repo_kind":"listed","path":"train_classifier_yelp.py","file_url":"https://github.com/ht1221/leap-lstm/blob/HEAD/train_classifier_yelp.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"287ebfa206000646"}},{"code_sha256_prefix":"26b33751bf044c3a","entry":"xavier_weights_variable","repo":"ht1221/leap-lstm","repo_kind":"listed","path":"skiplstm.py","file_url":"https://github.com/ht1221/leap-lstm/blob/HEAD/skiplstm.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"26b33751bf044c3a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}