{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/method/feedback-memory","slug":"feedback-memory","name":"Feedback Memory","full_name":"Feedback Memory","full_name_withheld":false,"description_markdown":"**Feedback Memory** is a type of attention module used in the [Feedback Transformer](https://paperswithcode.com/method/feedback-transformer) architecture. It allows a [transformer](https://paperswithcode.com/method/transformer) to to use the most abstract representations from the past directly as inputs for the current timestep. This means that the model does not form its representation in parallel, but sequentially token by token. More precisely, we replace the context inputs to attention modules with memory vectors that are computed over the past, i.e.:\r\n\r\n$$ \\mathbf{z}^{l}\\_{t} = \\text{Attn}\\left(\\mathbf{x}^{l}\\_{t},  \\left[\\mathbf{m}\\_{t−\\tau}, \\dots, \\mathbf{m}\\_{t−1}\\right]\\right) $$\r\n\r\nwhere a memory vector $\\mathbf{m}\\_{t}$ is computed by summing the representations of each layer at the $t$-th time step:\r\n\r\n$$ \\mathbf{m}\\_{t} = \\sum^{L}\\_{l=0}\\text{Softmax}\\left(w^{l}\\right)\\mathbf{x}\\_{t}^{l} $$\r\n\r\nwhere $w^{l}$ are learnable scalar parameters. Here $l = 0$ corresponds to token embeddings. The weighting of different layers by a [softmax](https://paperswithcode.com/method/softmax) output gives the model more flexibility as it can average them or select one of them. This modification of the self-attention input adapts the computation of the Transformer from parallel to sequential, summarized in the Figure. Indeed, it gives the ability to formulate the representation $\\mathbf{x}^{l}\\_{t+1}$ based on past representations from any layer $l'$, while in a standard Transformer this is only true for $l > l'$. This change can be viewed as exposing all previous computations to all future computations, providing better representations of the input. Such capacity would allow much shallower models to capture the same level of abstraction as a deeper architecture.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Addressing Some Limitations of Transformers with Feedback Memory","paper":"/paper/accessing-higher-level-representations-in","first_author":"Angela Fan","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/accessing-higher-level-representations-in"},"source":{"url":"https://arxiv.org/abs/2002.09402v3","title":"Addressing Some Limitations of Transformers with Feedback Memory","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Attention Modules","url":"/methods/category/attention-modules","pwc_aliases":[]}],"n_papers_tagged":4,"archive_num_papers":4,"papers_newest_first":[{"paper":null,"slug":"spire-boosting-llm-inference-throughput-with","title":"SPIRe: Boosting LLM Inference Throughput with Speculative Decoding","date":"2025-04-08","arxiv_id":"2504.06419","n_code_links":0,"syntology":null},{"paper":"/paper/iterative-feedback-network-for-unsupervised","slug":"iterative-feedback-network-for-unsupervised","title":"Iterative Feedback Network for Unsupervised Point Cloud Registration","date":"2024-01-09","arxiv_id":"2401.04357","n_code_links":1,"syntology":null},{"paper":"/paper/do-you-know-my-emotion-emotion-aware-strategy","slug":"do-you-know-my-emotion-emotion-aware-strategy","title":"Do You Know My Emotion? Emotion-Aware Strategy Recognition towards a Persuasive Dialogue System","date":"2022-06-24","arxiv_id":"2206.12101","n_code_links":1,"syntology":null},{"paper":"/paper/accessing-higher-level-representations-in","slug":"accessing-higher-level-representations-in","title":"Addressing Some Limitations of Transformers with Feedback Memory","date":"2020-02-21","arxiv_id":"2002.09402","n_code_links":4,"syntology":{"ran":3,"of":3,"n_ran_checked":2,"n_instrument":1,"unverified":0,"pointer_only":2,"phrase":"3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified","official":{"repos":["facebookresearch/transformer-sequential"],"state":"official: no sample here; runs from other or unrecorded repositories","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed","unlocated"]}}}],"papers_shown":4,"papers_order":"date (newest first), then slug","paper_counts":{"archive_papers_tagged":4,"with_a_code_link":3,"where_syntology_ran_a_sample":1,"not_listed_spam_title":0,"listed":4,"listed_where_code_ran":1,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":1,"every_run_a_failure_of_syntologys_instrument":0,"listed_with_a_run_with_no_instrument_failure":1,"listed_every_run_a_failure_of_syntologys_instrument":0,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers the archive tags; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"papers_where_code_ran":1,"papers_where_code_ran_first_page":"/method/feedback-memory/papers/ran/1","papers_where_code_ran_caption":"We ran code from the paper's repository; we did not isolate this method inside it.","pages":[{"order":"ran","page":1,"of":1,"url":"/method/feedback-memory/papers/ran/1","json":"/method/feedback-memory/papers/ran/1.json","rows":1}],"tasks":[{"task":"/task/language-modeling","name":"Language Modeling","papers":1},{"task":"/task/language-modelling","name":"Language Modelling","papers":1},{"task":"/task/machine-translation","name":"Machine Translation","papers":1},{"task":"/task/point-cloud-registration","name":"Point Cloud Registration","papers":1},{"task":"/task/reinforcement-learning","name":"Reinforcement Learning","papers":1},{"task":"/task/translation","name":"Translation","papers":1}],"tasks_shown":6,"n_tasks":6,"usage_by_year":[{"year":"2020","papers":1},{"year":"2022","papers":1},{"year":"2024","papers":1},{"year":"2025","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/feedback-memory"},"syntology_read_at":"2026-10-05T00:36:11+00:00","syntology":{"source":"Syntology's have check for this name (the free coverage lookup), read from Syntology's graph","have_url":"https://syntology.ai/api/graph/have?x=Feedback%20Memory","read_at":"2026-10-05T00:55:57+00:00","query":"Feedback Memory","kind":"method","have":"unknown_name","tier":null,"verified_implementation":null,"level":null,"level_meaning":"the served verification level, per answer; never summed across methods","resolved_to":null,"resolution":null,"resolved_name_differs":false,"absence_asserted":false,"absence_unchecked":null,"no_code_is_about_this_method":false,"subject":null,"archive_paper_arxiv_id":"2002.09402","repositories":null,"retrieve_with":null,"name_competition":{"verdict":"none","may_be_wrong_sense":false,"bearers_found":0,"served":null,"leading_other_bearer":null,"n_other_bearers_listed":0,"note":"Nothing in this corpus is named 'Feedback Memory' -- no method node, no paper announcing it in its title, no expanded form declaring it as an acronym."},"name_competition_text":null,"near_matches":["Distilling Feedback into Memory-as-a-Tool","Resource-Efficient Iterative LLM-Based NAS with Feedback Memory","Memory-R2","TS-Memory","Memory-QA","Memory-V2V","OCR-Memory","Memory Lobe"],"code_nearby":[],"code_nearby_caption":null,"origin_paper":null,"line":"name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.","level_legend":null},"syntology_have_state":"loaded","record_sha256":"69ee907a4f1c0e30c32a978f511baf958373995f5962d347cf43efe31156a395","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}