{"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/el-attention-memory-efficient-lossless","title":"EL-Attention: Memory Efficient Lossless Attention for Generation","arxiv_id":"2105.04779","date":"2021-05-11","proceeding":null,"authors":["Yu Yan","Jiusheng Chen","Weizhen Qi","Nikhil Bhendawade","Yeyun Gong","Nan Duan","Ruofei Zhang"],"abstract":"Transformer model with multi-head attention requires caching intermediate results for efficient inference in generation tasks. However, cache brings new memory-related costs and prevents leveraging larger batch size for faster speed. We propose memory-efficient lossless attention (called EL-attention) to address this issue. It avoids heavy operations for building multi-head keys and values, cache for them is not needed. EL-attention constructs an ensemble of attention results by expanding query while keeping key and value shared. It produces the same result as multi-head attention with less GPU memory and faster inference speed. We conduct extensive experiments on Transformer, BART, and GPT-2 for summarization and question generation tasks. The results show EL-attention speeds up existing models by 1.6x to 5.3x without accuracy loss.","url_abs":"https://arxiv.org/abs/2105.04779v2","url_pdf":"https://arxiv.org/pdf/2105.04779v2.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":"el-attention-memory-efficient-lossless","repo_url":"https://github.com/microsoft/fastseq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":null,"task_name":"GPU"},{"task_slug":"question-generation","task_name":"Question Generation"},{"task_slug":"question-generation","task_name":"Question-Generation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bart","method_name":"BART"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"discriminative-fine-tuning","method_name":"Discriminative Fine-Tuning"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-2","method_name":"GPT-2"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"weight-decay","method_name":"Weight Decay"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2105.04779","atlas_url":"https://app.syntology.ai/?focus=2105.04779","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2105.04779"}},"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/microsoft/fastseq","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"unverified":4},"by_repo_kind":{"official":{"samples":6,"ran":2,"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":"96cbb5e9ca5b6be0","entry":"Embedding","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq/models/prophetnet_fs/ngram_s2s_model.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq/models/prophetnet_fs/ngram_s2s_model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"96cbb5e9ca5b6be0"}},{"code_sha256_prefix":"8cd8cb0d1e9e63e4","entry":"Linear","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq/models/prophetnet_fs/ngram_s2s_model.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq/models/prophetnet_fs/ngram_s2s_model.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8cd8cb0d1e9e63e4"}},{"code_sha256_prefix":"314b4accae5bf9b5","entry":"encode_line","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq_cli/transformers_utils.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq_cli/transformers_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"314b4accae5bf9b5"}},{"code_sha256_prefix":"e386d3a24af4d168","entry":"label_smoothed_nll_loss","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq_cli/transformers_utils.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq_cli/transformers_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e386d3a24af4d168"}},{"code_sha256_prefix":"6a66f22f0034f9d3","entry":"lmap","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq_cli/transformers_utils.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq_cli/transformers_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6a66f22f0034f9d3"}},{"code_sha256_prefix":"c2d34f90558f3cf6","entry":"ngram_attention_bias","repo":"microsoft/fastseq","repo_kind":"official","path":"fastseq/models/prophetnet_fs/ngram_multihead_attention.py","file_url":"https://github.com/microsoft/fastseq/blob/HEAD/fastseq/models/prophetnet_fs/ngram_multihead_attention.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c2d34f90558f3cf6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}