{"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":"/paper/towards-economical-inference-enabling","title":"Towards Economical Inference: Enabling DeepSeek's Multi-Head Latent Attention in Any Transformer-based LLMs","arxiv_id":"2502.14837","date":"2025-02-20","proceeding":null,"authors":["Tao Ji","Bin Guo","Yuanbin Wu","Qipeng Guo","Lixing Shen","Zhan Chen","Xipeng Qiu","Qi Zhang","Tao Gui"],"abstract":"Multi-head Latent Attention (MLA) is an innovative architecture proposed by DeepSeek, designed to ensure efficient and economical inference by significantly compressing the Key-Value (KV) cache into a latent vector. Compared to MLA, standard LLMs employing Multi-Head Attention (MHA) and its variants such as Grouped-Query Attention (GQA) exhibit significant cost disadvantages. Enabling well-trained LLMs (e.g., Llama) to rapidly adapt to MLA without pre-training from scratch is both meaningful and challenging. This paper proposes the first data-efficient fine-tuning method for transitioning from MHA to MLA (MHA2MLA), which includes two key components: for partial-RoPE, we remove RoPE from dimensions of queries and keys that contribute less to the attention scores, for low-rank approximation, we introduce joint SVD approximations based on the pre-trained parameters of keys and values. These carefully designed strategies enable MHA2MLA to recover performance using only a small fraction (0.3% to 0.6%) of the data, significantly reducing inference costs while seamlessly integrating with compression techniques such as KV cache quantization. For example, the KV cache size of Llama2-7B is reduced by 92.19%, with only a 0.5% drop in LongBench performance.","url_abs":"https://arxiv.org/abs/2502.14837v1","url_pdf":"https://arxiv.org/pdf/2502.14837v1.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":"towards-economical-inference-enabling","repo_url":"https://github.com/JT-Ushio/MHA2MLA","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"quantization","task_name":"Quantization"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"feedforward-network","method_name":"Feedforward Network"},{"method_slug":"grouped-query-attention","method_name":"Grouped-query attention"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2502.14837","atlas_url":"https://app.syntology.ai/?focus=2502.14837","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2502.14837"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2502.14837"},"returns":"the samples harvested from this paper, each with its run record; it does not verify the paper"},{"tool":"get_code","arguments_in":"samples[].code_sha256","returns":"one stored sample by its code_sha256; not shown on this site"}]},"developers":"https://syntology.ai/developers","held":true,"reason":"in_graph_layer","checked_against":null,"agent_calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2502.14837"},"call":"get_harvested_code_for_paper(arxiv_id=\"2502.14837\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2502.14837"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2502.14837\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2502.14837"},"call":"get_concepts_for_paper(arxiv_id=\"2502.14837\")"}],"by_repo":[{"repo":"JT-Ushio/MHA2MLA","kind":"official","samples":15,"ran":10,"instrument":0,"unverified":5}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code","read_at":"2026-10-05T00:36:11+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","ran_record":{"url":"https://syntology.ai/api/ran/2502.14837.json","schema":"syntology.ran_record/1","counts_crosswalk":{"n_samples":"syntology.counts.lifted","n_ran":"syntology.counts.executed","n_ran_checked":"syntology.counts.checked","n_instrument":"syntology.counts.instrument_failures","n_unverified":"syntology.counts.no_recorded_run"},"note":"live record, measured at request time; may be newer than this page's graph read","badge":"https://syntology.ai/api/ran/2502.14837.svg"},"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/JT-Ushio/MHA2MLA","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":10,"unverified":5},"n_samples":15,"n_ran":10,"n_constructed":0,"n_ran_checked":10,"n_instrument":0,"n_unverified":5,"n_instrument_is":"failures of Syntology's instrument, not of the code","every_run_is_an_instrument_failure":false,"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"by_repo_kind":{"official":{"samples":15,"ran":10,"n_ran_checked":10,"n_instrument":0,"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,"inline_ok_is":"inline_ok is true only when both licence gates allow Syntology to serve this code's text: this paper's link (inline_ok_link, from the licence of this paper's copy) and the stored copy of the same code that Syntology serves, which may have been first harvested from another repository. withheld_clause is Syntology's own wording of which gate keeps the text a pointer and why, and is null when inline_ok is true. This site shows no code text either way; n_pointer_only_for_licence counts samples with inline_ok false (both gates), n_pointer_only_link_allows those of them whose inline_ok_link is true","n_pointer_only_link_allows":0,"fixture_fields_is":{"signature_and_fixture_out_type":"the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right","level_recorded_at":"when Syntology's graph wrote the sample's verification level, not when the code executed; absent where the level was written before the date was kept, and never filled in with the build date","present_on":"samples that ran only; a key is absent, never empty, where the graph records nothing"},"n_ran_with_level_date":10,"n_ran_without_level_date":0,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-24","last":"2026-09-25","is":"the first and last UTC dates on which Syntology's graph wrote a verification level for a sample on this page that ran, not when the code executed"},"samples":[{"code_sha256":"d1722a47a6e68260","code_sha256_prefix":"d1722a47a6e68260","entry":"SVD","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patch_func.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patch_func.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"X:Tensor/2/float, r:int/0/int","fixture_out_type":"tuple","level_recorded_at":"2026-09-25T04:00:33.621019+00:00","retrieve_with":{"tool":"get_code","code_sha256":"d1722a47a6e68260"},"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"d1722a47a6e68260"}},{"code_sha256":"bd8982c35600a2b7","code_sha256_prefix":"bd8982c35600a2b7","entry":"create_custom_apply_rotary_pos_emb","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_llama.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_llama.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"q_r_indices:list/1/Tensor, k_r_indices:list/1/Tensor","fixture_out_type":"function","level_recorded_at":"2026-09-24T23:57:41.519889+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"bd8982c35600a2b7"}},{"code_sha256":"f73327d915050f1a","code_sha256_prefix":"f73327d915050f1a","entry":"create_custom_apply_rotary_pos_emb","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen2.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen2.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"q_r_indices:list/1/Tensor, k_r_indices:list/1/Tensor","fixture_out_type":"function","level_recorded_at":"2026-09-24T23:57:41.519889+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"f73327d915050f1a"}},{"code_sha256":"4b3aa5813125a737","code_sha256_prefix":"4b3aa5813125a737","entry":"create_custom_apply_rotary_pos_emb","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen3.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen3.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"q_r_indices:list/1/Tensor, k_r_indices:list/1/Tensor","fixture_out_type":"function","level_recorded_at":"2026-09-25T04:00:33.621019+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"4b3aa5813125a737"}},{"code_sha256":"e3be3d316dce65ea","code_sha256_prefix":"e3be3d316dce65ea","entry":"create_custom_apply_rotary_pos_emb","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen3_moe.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen3_moe.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"q_r_indices:list/1/Tensor, k_r_indices:list/1/Tensor","fixture_out_type":"function","level_recorded_at":"2026-09-24T23:57:41.519889+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"e3be3d316dce65ea"}},{"code_sha256":"3e838680cc326966","code_sha256_prefix":"3e838680cc326966","entry":"create_hook_fn","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/2_norm.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/2_norm.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"name:str/0/str","fixture_out_type":"function","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"3e838680cc326966"}},{"code_sha256":"c72933732661bbe4","code_sha256_prefix":"c72933732661bbe4","entry":"generate_random_input_ids","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"eval/benchmark_latency_memory.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/eval/benchmark_latency_memory.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"batch_size:int/0/int, input_length:int/0/int, vocab_size:int/0/int, device:str/0/str","fixture_out_type":"Tensor","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"c72933732661bbe4"}},{"code_sha256":"818485f3147eb543","code_sha256_prefix":"818485f3147eb543","entry":"partial_rope_mask","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patch_func.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patch_func.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"model_args:ModelArgs/0/?, mha2mla_args:MHA2MLAArgs/0/?","fixture_out_type":"tuple","level_recorded_at":"2026-09-24T23:57:41.519889+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"818485f3147eb543"}},{"code_sha256":"937f754d0af3a99a","code_sha256_prefix":"937f754d0af3a99a","entry":"reorder_matrix_rows","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_model_load.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_model_load.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"mask:Tensor/1/bool, is_cat:bool/0/bool","fixture_out_type":"tuple","level_recorded_at":"2026-09-24T23:57:41.519889+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"937f754d0af3a99a"}},{"code_sha256":"1b387fd9635cb3ad","code_sha256_prefix":"1b387fd9635cb3ad","entry":"svd_low_rank_approx","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patch_func.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patch_func.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"k_c_weight:Tensor/2/float, k_c_bias:Tensor/1/float, v_weight:Tensor/2/float, v_bias:Tensor/1/float, d_kv_mid:int/0/int, method:str/0/str","fixture_out_type":"LowRankKVLinear","level_recorded_at":"2026-09-24T23:25:36.931776+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"1b387fd9635cb3ad"}},{"code_sha256":"19448606f10d10d8","code_sha256_prefix":"19448606f10d10d8","entry":"custom_LlamaAttention_forward","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_llama.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_llama.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"19448606f10d10d8"}},{"code_sha256":"7ac64d7242a38034","code_sha256_prefix":"7ac64d7242a38034","entry":"custom_Qwen2Attention_forward","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen2.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen2.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"7ac64d7242a38034"}},{"code_sha256":"ccace102b231da7d","code_sha256_prefix":"ccace102b231da7d","entry":"custom_Qwen3Attention_forward","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen3.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen3.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"ccace102b231da7d"}},{"code_sha256":"88a635fd8ae25685","code_sha256_prefix":"88a635fd8ae25685","entry":"custom_Qwen3MoeAttention_forward","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/patching_qwen3_moe.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/patching_qwen3_moe.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"88a635fd8ae25685"}},{"code_sha256":"c836c0fa19b74615","code_sha256_prefix":"c836c0fa19b74615","entry":"mha2mla_ref","repo":"JT-Ushio/MHA2MLA","repo_kind":"official","path":"src/mha2mla/mla_triton_kernel.py","file_url":"https://github.com/JT-Ushio/MHA2MLA/blob/HEAD/src/mha2mla/mla_triton_kernel.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"c836c0fa19b74615"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"73f6b0bc83d5629b0b213cbea6122eb4f05822cb722a7b6e6b264eae61a0b9b0","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}