{"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/an-evolved-universal-transformer-memory","title":"An Evolved Universal Transformer Memory","arxiv_id":"2410.13166","date":"2024-10-17","proceeding":null,"authors":["Edoardo Cetin","Qi Sun","Tianyu Zhao","Yujin Tang"],"abstract":"Prior methods propose to offset the escalating costs of modern foundation models by dropping specific parts of their contexts with hand-designed rules, while attempting to preserve their original performance. We overcome this trade-off with Neural Attention Memory Models (NAMMs), introducing a learned network for memory management that improves both the performance and efficiency of transformers. We evolve NAMMs atop pre-trained transformers to provide different latent contexts focusing on the most relevant information for individual layers and attention heads. NAMMs are universally applicable to any model using self-attention as they condition exclusively on the values in the produced attention matrices. Learning NAMMs on a small set of problems, we achieve substantial performance improvements across multiple long-context benchmarks while cutting the model's input contexts up to a fraction of the original sizes. We show the generality of our conditioning enables zero-shot transfer of NAMMs trained only on language to entirely new transformer architectures even across input modalities, with their benefits carrying over to vision and reinforcement learning.","url_abs":"https://arxiv.org/abs/2410.13166v3","url_pdf":"https://arxiv.org/pdf/2410.13166v3.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":"an-evolved-universal-transformer-memory","repo_url":"https://github.com/sakanaai/evo-memory","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"set","method_name":"SET"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.13166","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.13166"}},"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":"deterministic:regex_extraction","url":"https://github.com/SakanaAI/evo-memory","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/sakanaai/evo-memory","reach":{"status":"ok"}}],"summary":{"ran":11,"unverified":1},"by_repo_kind":{"official":{"samples":12,"ran":11,"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":12,"samples":[{"code_sha256_prefix":"8979b56a0788125e","entry":"compute_recency","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_dynamic.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_dynamic.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8979b56a0788125e"}},{"code_sha256_prefix":"b138114d7116560c","entry":"compute_recency_scores","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_dynamic.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_dynamic.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b138114d7116560c"}},{"code_sha256_prefix":"aa010c2bd0aa74b0","entry":"convert_to_tensor","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/deep_embedding.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/deep_embedding.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"aa010c2bd0aa74b0"}},{"code_sha256_prefix":"a6c6ee847acad66b","entry":"get_cma_defaults","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_evolution/cma_es.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_evolution/cma_es.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a6c6ee847acad66b"}},{"code_sha256_prefix":"36754005e03e2458","entry":"get_property_list_from_elements","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_deep_components.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_deep_components.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"36754005e03e2458"}},{"code_sha256_prefix":"82650da99214bdf2","entry":"reduce_stats","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/deep.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/deep.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"82650da99214bdf2"}},{"code_sha256_prefix":"941a06561780b35a","entry":"stat_fn_list","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/deep.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/deep.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"941a06561780b35a"}},{"code_sha256_prefix":"ed9673e65d4694aa","entry":"stat_fn_tensor","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/deep.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/deep.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ed9673e65d4694aa"}},{"code_sha256_prefix":"5033508c5b925cb2","entry":"threshold_score_idxs","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_dynamic.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_dynamic.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5033508c5b925cb2"}},{"code_sha256_prefix":"998a8e8e5f062104","entry":"true_for_all","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_deep_components.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_deep_components.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"998a8e8e5f062104"}},{"code_sha256_prefix":"62e579d3352bdb8e","entry":"true_for_any","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_policy/base_deep_components.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_policy/base_deep_components.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"62e579d3352bdb8e"}},{"code_sha256_prefix":"0c0fa69e90becc08","entry":"full_eigen_decomp","repo":"SakanaAI/evo-memory","repo_kind":"official","path":"memory_evolution/cma_es.py","file_url":"https://github.com/SakanaAI/evo-memory/blob/HEAD/memory_evolution/cma_es.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0c0fa69e90becc08"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}