{"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/deep-equilibrium-models","title":"Deep Equilibrium Models","arxiv_id":"1909.01377","date":"2019-09-03","proceeding":"NeurIPS 2019 12","authors":["Shaojie Bai","J. Zico Kolter","Vladlen Koltun"],"abstract":"We present a new approach to modeling sequential data: the deep equilibrium model (DEQ). Motivated by an observation that the hidden layers of many existing deep sequence models converge towards some fixed point, we propose the DEQ approach that directly finds these equilibrium points via root-finding. Such a method is equivalent to running an infinite depth (weight-tied) feedforward network, but has the notable advantage that we can analytically backpropagate through the equilibrium point using implicit differentiation. Using this approach, training and prediction in these networks require only constant memory, regardless of the effective \"depth\" of the network. We demonstrate how DEQs can be applied to two state-of-the-art deep sequence models: self-attention transformers and trellis networks. On large-scale language modeling tasks, such as the WikiText-103 benchmark, we show that DEQs 1) often improve performance over these state-of-the-art models (for similar parameter counts); 2) have similar computational requirements to existing models; and 3) vastly reduce memory consumption (often the bottleneck for training large sequence models), demonstrating an up-to 88% memory reduction in our experiments. The code is available at https://github.com/locuslab/deq .","url_abs":"https://arxiv.org/abs/1909.01377v2","url_pdf":"https://arxiv.org/pdf/1909.01377v2.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":"deep-equilibrium-models","repo_url":"https://github.com/locuslab/deq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/SinclairHudson/DeepEquilibrium","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/cgoemaere/hamdeq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/cgoemaere/hopdeq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/locuslab/impsq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/locuslab/monotone_op_net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/lufanma/ifr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/martaskrt/qdeq","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/prolearner/hypertorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/reacho/deep-equilibrium-vs-bilevel","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"deep-equilibrium-models","repo_url":"https://github.com/sciml/fastdeq.jl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"deq","method_name":"DEQ"}],"datasets_introduced":[],"methods_introduced":[{"slug":"deq","name":"DEQ","full_name":"Deep Equilibrium Models"}],"results":[{"leaderboard":"/sota/language-modelling-on-penn-treebank-word","task":"Language Modelling","dataset":"Penn Treebank (Word Level)","model":"DEQ-TrellisNet","rank_in_archive_order":29,"of":43,"metrics":{"Params":"24M","Test perplexity":"57.1"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-wikitext-103","task":"Language Modelling","dataset":"WikiText-103","model":"DEQ-Transformer (medium, adaptive embed)","rank_in_archive_order":50,"of":89,"metrics":{"Number of params":"110M","Test perplexity":"23.2"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-wikitext-103","task":"Language Modelling","dataset":"WikiText-103","model":"DEQ-TrellisNet","rank_in_archive_order":66,"of":89,"metrics":{"Number of params":"180M","Test perplexity":"29.0"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-wikitext-103","task":"Language Modelling","dataset":"WikiText-103","model":"DEQ-Transformer (small)","rank_in_archive_order":73,"of":89,"metrics":{"Number of params":"138M","Test perplexity":"32.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1909.01377","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1909.01377"}},"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/locuslab/monotone_op_net","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/prolearner/hypertorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/martaskrt/qdeq","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/locuslab/impsq","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cgoemaere/hamdeq","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/reacho/deep-equilibrium-vs-bilevel","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/locuslab/deq","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lufanma/ifr","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/cgoemaere/hopdeq","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/SinclairHudson/DeepEquilibrium","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sciml/fastdeq.jl","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"ran_honours":1,"unverified":9},"by_repo_kind":{"listed":{"samples":11,"ran":3,"repositories":3}},"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":4,"samples":[{"code_sha256_prefix":"635c8fdeade67cba","entry":"load_dataset","repo":"locuslab/impsq","repo_kind":"listed","path":"scripts/train_2d_image.py","file_url":"https://github.com/locuslab/impsq/blob/HEAD/scripts/train_2d_image.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"635c8fdeade67cba"}},{"code_sha256_prefix":"81a8d0038b2d548a","entry":"postprocess","repo":"locuslab/impsq","repo_kind":"listed","path":"scripts/train_2d_image.py","file_url":"https://github.com/locuslab/impsq/blob/HEAD/scripts/train_2d_image.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"81a8d0038b2d548a"}},{"code_sha256_prefix":"1194fc2f2f213578","entry":"preprocess","repo":"locuslab/impsq","repo_kind":"listed","path":"scripts/train_2d_image.py","file_url":"https://github.com/locuslab/impsq/blob/HEAD/scripts/train_2d_image.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"1194fc2f2f213578"}},{"code_sha256_prefix":"799e823c70a9d96b","entry":"anderson","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/solvers.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/solvers.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":"799e823c70a9d96b"}},{"code_sha256_prefix":"d9164060df059089","entry":"backprop","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/backward.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/backward.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":"d9164060df059089"}},{"code_sha256_prefix":"1ec688ed87b414c9","entry":"damped","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/utils/fixed_point_utils.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/utils/fixed_point_utils.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":"1ec688ed87b414c9"}},{"code_sha256_prefix":"ee0aeb0dbfe062f7","entry":"evaluate","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"ee0aeb0dbfe062f7"}},{"code_sha256_prefix":"8b3921143c779266","entry":"full_adjoint","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/backward.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/backward.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":"8b3921143c779266"}},{"code_sha256_prefix":"06ee28cae377b42a","entry":"get_function_from_package","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/utils/import_utils.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/utils/import_utils.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":"06ee28cae377b42a"}},{"code_sha256_prefix":"a0f601462ac10944","entry":"picard","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/solvers.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/solvers.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":"a0f601462ac10944"}},{"code_sha256_prefix":"20e21defecd44b23","entry":"power_iteration_torch_T","repo":"reacho/deep-equilibrium-vs-bilevel","repo_kind":"listed","path":"CelebA/DEQ-vs-bilevel-conv2d.py","file_url":"https://github.com/reacho/deep-equilibrium-vs-bilevel/blob/HEAD/CelebA/DEQ-vs-bilevel-conv2d.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":"20e21defecd44b23"}},{"code_sha256_prefix":"641ed886ac5440c1","entry":"track_states","repo":"cgoemaere/hamdeq","repo_kind":"listed","path":"hopdeq/deq_core/utils/fixed_point_utils.py","file_url":"https://github.com/cgoemaere/hamdeq/blob/HEAD/hopdeq/deq_core/utils/fixed_point_utils.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":"641ed886ac5440c1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}