{"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/tofu","slug":"tofu","name":"Tofu","full_name":"Tofu","full_name_withheld":false,"description_markdown":"**Tofu** is an intra-layer model parallel system that partitions very large DNN models across multiple GPU devices to reduce per-GPU memory footprint. Tofu is designed to partition a dataflow graph of fine-grained tensor operators used by platforms like MXNet and TensorFlow. To optimally partition different operators in a dataflow graph, Tofu uses a recursive search algorithm that minimizes the total communication cost.","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":"http://arxiv.org/abs/1807.08887v2","title":"Supporting Very Large Models using Automatic Dataflow Graph Partitioning","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":"Distributed Methods","url":"/methods/category/distributed-methods","pwc_aliases":[]}],"n_papers_tagged":19,"archive_num_papers":null,"papers_newest_first":[{"paper":null,"slug":"guard-guided-unlearning-and-retention-via","title":"GUARD: Guided Unlearning and Retention via Data Attribution for Large Language Models","date":"2025-06-12","arxiv_id":"2506.10946","n_code_links":0,"syntology":null},{"paper":null,"slug":"constrained-entropic-unlearning-a-primal-dual","title":"Constrained Entropic Unlearning: A Primal-Dual Framework for Large Language Models","date":"2025-06-05","arxiv_id":"2506.05314","n_code_links":0,"syntology":null},{"paper":"/paper/unierase-unlearning-token-as-a-universal","slug":"unierase-unlearning-token-as-a-universal","title":"UniErase: Unlearning Token as a Universal Erasure Primitive for Language Models","date":"2025-05-21","arxiv_id":"2505.15674","n_code_links":1,"syntology":null},{"paper":null,"slug":"guard-generation-time-llm-unlearning-via","title":"GUARD: Generation-time LLM Unlearning via Adaptive Restriction and Detection","date":"2025-05-19","arxiv_id":"2505.13312","n_code_links":0,"syntology":null},{"paper":null,"slug":"uipe-enhancing-llm-unlearning-by-removing","title":"UIPE: Enhancing LLM Unlearning by Removing Knowledge Related to Forgetting Targets","date":"2025-03-06","arxiv_id":"2503.04693","n_code_links":0,"syntology":null},{"paper":null,"slug":"ce-u-cross-entropy-unlearning","title":"CE-U: Cross Entropy Unlearning","date":"2025-03-03","arxiv_id":"2503.01224","n_code_links":0,"syntology":null},{"paper":"/paper/towards-robust-evaluation-of-unlearning-in","slug":"towards-robust-evaluation-of-unlearning-in","title":"Towards Robust Evaluation of Unlearning in LLMs via Data Transformations","date":"2024-11-23","arxiv_id":"2411.15477","n_code_links":1,"syntology":null},{"paper":null,"slug":"unlearning-as-multi-task-optimization-a","title":"Unlearning as multi-task optimization: A normalized gradient difference approach with an adaptive learning rate","date":"2024-10-29","arxiv_id":"2410.22086","n_code_links":0,"syntology":null},{"paper":null,"slug":"llm-unlearning-via-loss-adjustment-with-only","title":"LLM Unlearning via Loss Adjustment with Only Forget Data","date":"2024-10-14","arxiv_id":"2410.11143","n_code_links":0,"syntology":null},{"paper":"/paper/simplicity-prevails-rethinking-negative","slug":"simplicity-prevails-rethinking-negative","title":"Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning","date":"2024-10-09","arxiv_id":"2410.07163","n_code_links":2,"syntology":{"ran":8,"of":10,"n_ran_checked":5,"n_instrument":3,"unverified":2,"pointer_only":1,"phrase":"8 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified","official":{"repos":["OPTML-Group/Unlearn-Simple"],"state":"official (archive's flag): 8 ran","n_ran":8,"n_constructed":0,"n_ran_no_instrument_failure":5,"n_unverified":2,"ran_from_kinds":["official"]}}},{"paper":null,"slug":"answer-when-needed-forget-when-not-language","title":"Answer When Needed, Forget When Not: Language Models Pretend to Forget via In-Context Knowledge Unlearning","date":"2024-10-01","arxiv_id":"2410.00382","n_code_links":0,"syntology":null},{"paper":"/paper/towards-robust-and-cost-efficient-knowledge","slug":"towards-robust-and-cost-efficient-knowledge","title":"Towards Robust and Parameter-Efficient Knowledge Unlearning for LLMs","date":"2024-08-13","arxiv_id":"2408.06621","n_code_links":1,"syntology":{"ran":2,"of":2,"n_ran_checked":0,"n_instrument":2,"unverified":0,"pointer_only":0,"phrase":"2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified","official":{"repos":["csm9493/efficient-llm-unlearning"],"state":"official (archive's flag): 2 ran","n_ran":2,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["official"]}}},{"paper":"/paper/reversing-the-forget-retain-objectives-an","slug":"reversing-the-forget-retain-objectives-an","title":"Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference","date":"2024-06-12","arxiv_id":"2406.08607","n_code_links":1,"syntology":null},{"paper":"/paper/negative-preference-optimization-from","slug":"negative-preference-optimization-from","title":"Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning","date":"2024-04-08","arxiv_id":"2404.05868","n_code_links":1,"syntology":null},{"paper":null,"slug":"token-fusion-bridging-the-gap-between-token","title":"Token Fusion: Bridging the Gap between Token Pruning and Token Merging","date":"2023-12-02","arxiv_id":"2312.01026","n_code_links":0,"syntology":null},{"paper":null,"slug":"on-high-dimensional-and-low-rank-tensor","title":"On High-dimensional and Low-rank Tensor Bandits","date":"2023-05-06","arxiv_id":"2305.03884","n_code_links":0,"syntology":null},{"paper":null,"slug":"tofu-towards-obfuscated-federated-updates-by","title":"TOFU: Towards Obfuscated Federated Updates by Encoding Weight Updates into Gradients from Proxy Data","date":"2022-01-21","arxiv_id":"2201.08494","n_code_links":0,"syntology":null},{"paper":null,"slug":"topologically-consistent-multi-view-face-1","title":"Topologically Consistent Multi-View Face Inference Using Volumetric Sampling","date":"2021-10-06","arxiv_id":"2110.02948","n_code_links":0,"syntology":null},{"paper":null,"slug":"supporting-very-large-models-using-automatic","title":"Supporting Very Large Models using Automatic Dataflow Graph Partitioning","date":"2018-07-24","arxiv_id":"1807.08887","n_code_links":0,"syntology":null}],"papers_shown":19,"papers_order":"date (newest first), then slug","paper_counts":{"archive_papers_tagged":19,"with_a_code_link":6,"where_syntology_ran_a_sample":2,"not_listed_spam_title":0,"listed":19,"listed_where_code_ran":2,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":1,"every_run_a_failure_of_syntologys_instrument":1,"listed_with_a_run_with_no_instrument_failure":1,"listed_every_run_a_failure_of_syntologys_instrument":1,"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":2,"papers_where_code_ran_first_page":"/method/tofu/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/tofu/papers/ran/1","json":"/method/tofu/papers/ran/1.json","rows":2}],"tasks":[{"task":"/task/machine-unlearning","name":"Machine Unlearning","papers":4},{"task":"/task/text-generation","name":"Text Generation","papers":2},{"task":"/task/3d-reconstruction","name":"3D Reconstruction","papers":1},{"task":"/task/computational-efficiency","name":"Computational Efficiency","papers":1},{"task":"/task/federated-learning","name":"Federated Learning","papers":1},{"task":null,"name":"GPU","papers":1},{"task":"/task/image-generation","name":"Image Generation","papers":1},{"task":"/task/language-modeling","name":"Language Modeling","papers":1},{"task":"/task/language-modelling","name":"Language Modelling","papers":1},{"task":"/task/large-language-model","name":"Large Language Model","papers":1},{"task":"/task/memorization","name":"Memorization","papers":1},{"task":"/task/model-editing","name":"Model Editing","papers":1},{"task":"/task/recommendation-systems","name":"Recommendation Systems","papers":1},{"task":"/task/reinforcement-learning","name":"Reinforcement Learning","papers":1},{"task":"/task/high","name":"Vocal Bursts Intensity Prediction","papers":1},{"task":"/task/world-knowledge","name":"World Knowledge","papers":1},{"task":"/task/graph-partitioning","name":"graph partitioning","papers":1},{"task":"/task/reinforcement-learning-2","name":"reinforcement-learning","papers":1}],"tasks_shown":18,"n_tasks":18,"usage_by_year":[{"year":"2018","papers":1},{"year":"2021","papers":1},{"year":"2022","papers":1},{"year":"2023","papers":2},{"year":"2024","papers":8},{"year":"2025","papers":6}],"row_source":"embedded","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/tofu"},"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=Tofu","read_at":"2026-10-05T01:03:08+00:00","query":"Tofu","kind":"method","have":"declared_code","tier":"repo_known","verified_implementation":false,"level":null,"level_meaning":"the served verification level, per answer; never summed across methods","resolved_to":{"method":"TOFU","origin_arxiv_id":"2607.11423"},"resolution":"case_insensitive","resolved_name_differs":false,"absence_asserted":false,"absence_unchecked":null,"no_code_is_about_this_method":false,"subject":"unknown","archive_paper_arxiv_id":null,"repositories":{"listed":2,"in_graph":2,"list_truncated":false,"strongest_link":"listed_only"},"retrieve_with":null,"name_competition":{"verdict":"outranked","may_be_wrong_sense":true,"bearers_found":7,"served":{"arxiv_id":"2607.11423","citations":0},"leading_other_bearer":{"arxiv_id":"2401.06121","title":"TOFU: A Task of Fictitious Unlearning for LLMs","citations":554,"named_by":["title_prefix"],"bearer_name":"TOFU: A Task of Fictitious Unlearning for LLMs","ask_with":"have('2401.06121')"},"n_other_bearers_listed":5,"note":"'Tofu' is ALSO the name of the paper 'TOFU: A Task of Fictitious Unlearning for LLMs' (arXiv:2401.06121, 554 citations), which is cited FAR more than the paper this answer is about (0). This answer may be about the wrong Tofu. If you meant the better-known one, ask again with that entry of `also_named`."},"name_competition_text":"The same name also belongs to the paper “TOFU: A Task of Fictitious Unlearning for LLMs” (arXiv:2401.06121, 554 citations), which the lookup finds cited far more than the paper this answer is about (0 citations), so this answer may be about a different sense of the name.","near_matches":null,"code_nearby":[],"code_nearby_caption":null,"origin_paper":{"arxiv_id":"2607.11423","page":"/paper/arxiv-2607-11423","syntology_url":"https://syntology.ai/paper/2607.11423","in_graph_layer":true,"samples":44,"ran":27,"n_ran_checked":6,"n_instrument":21,"unverified":17,"phrase":"27 ran (of which 4 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 1 violated, 4 with no contract checked; 21 where Syntology's instrument failed) · 17 unverified","instrument_means":"failures of Syntology's instrument, not of the code"},"line":"repository known, no verified implementation (declared_code; repositories: 2 listed of 2 in the graph, strongest link: listed only). The lookup matched 'TOFU' ignoring case. The archive credits no paper with an arXiv id for this method, so whether this answer (about arXiv:2607.11423) is about this method is not checked.","level_legend":null},"syntology_have_state":"loaded","record_sha256":"dca6ffd5b778abe66fedbefca90583d4262eae077d498cfc8f209ec2f902a2ae","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}