{"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/tensor-product-attention-is-all-you-need","title":"Tensor Product Attention Is All You Need","arxiv_id":"2501.06425","date":"2025-01-11","proceeding":null,"authors":["Yifan Zhang","Yifeng Liu","Huizhuo Yuan","Zhen Qin","Yang Yuan","Quanquan Gu","Andrew Chi-Chih Yao"],"abstract":"Scaling language models to handle longer input sequences typically necessitates large key-value (KV) caches, resulting in substantial memory overhead during inference. In this paper, we propose Tensor Product Attention (TPA), a novel attention mechanism that uses tensor decompositions to represent queries, keys, and values compactly, significantly shrinking KV cache size at inference time. By factorizing these representations into contextual low-rank components (contextual factorization) and seamlessly integrating with RoPE, TPA achieves improved model quality alongside memory efficiency. Based on TPA, we introduce the Tensor ProducT ATTenTion Transformer (T6), a new model architecture for sequence modeling. Through extensive empirical evaluation of language modeling tasks, we demonstrate that T6 exceeds the performance of standard Transformer baselines including MHA, MQA, GQA, and MLA across various metrics, including perplexity and a range of renowned evaluation benchmarks. Notably, TPAs memory efficiency enables the processing of significantly longer sequences under fixed resource constraints, addressing a critical scalability challenge in modern language models. The code is available at https://github.com/tensorgi/T6.","url_abs":"https://arxiv.org/abs/2501.06425v1","url_pdf":"https://arxiv.org/pdf/2501.06425v1.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":"tensor-product-attention-is-all-you-need","repo_url":"https://github.com/tensorgi/t6","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"all","task_name":"All"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2501.06425","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2501.06425"}},"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/tensorgi/t6","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":4,"ran_violates":1,"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":6,"ran":3,"repositories":1},"community":{"samples":3,"ran":3,"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,"samples":[{"code_sha256_prefix":"b47d48e431b34acd","entry":"apply_rotary_emb","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6_infer.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6_infer.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b47d48e431b34acd"}},{"code_sha256_prefix":"46c4b171da8b3c7a","entry":"flashtpa_decode_torch","repo":"tensorgi/tpa","repo_kind":"community","path":"decode/flashtpa_decode_torch.py","file_url":"https://github.com/tensorgi/tpa/blob/HEAD/decode/flashtpa_decode_torch.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"46c4b171da8b3c7a"}},{"code_sha256_prefix":"e93cc5b705c2eb3b","entry":"precompute_freqs_cis","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6_infer.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6_infer.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e93cc5b705c2eb3b"}},{"code_sha256_prefix":"c44830a7722ff5f7","entry":"request_caching_arg_to_dict","repo":"tensorgi/TPA","repo_kind":"community","path":"lm-evaluation-harness/lm_eval/evaluator.py","file_url":"https://github.com/tensorgi/TPA/blob/HEAD/lm-evaluation-harness/lm_eval/evaluator.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c44830a7722ff5f7"}},{"code_sha256_prefix":"70bf6ebaafd266c4","entry":"reshape_for_broadcast","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6_infer.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6_infer.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"70bf6ebaafd266c4"}},{"code_sha256_prefix":"05f139bc4395fbd8","entry":"tpa_decode_naive_torch","repo":"tensorgi/tpa","repo_kind":"community","path":"decode/flashtpa_decode_torch.py","file_url":"https://github.com/tensorgi/tpa/blob/HEAD/decode/flashtpa_decode_torch.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"05f139bc4395fbd8"}},{"code_sha256_prefix":"0f4f2333e89a6c8c","entry":"apply_rotary_emb","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6.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":"0f4f2333e89a6c8c"}},{"code_sha256_prefix":"c2a8d2ee1e68a6db","entry":"apply_rotary_emb","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6_kvonly_partialrope.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6_kvonly_partialrope.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":"c2a8d2ee1e68a6db"}},{"code_sha256_prefix":"31eead4c21084b8b","entry":"apply_rotary_emb","repo":"tensorgi/t6","repo_kind":"official","path":"model/T6_linearattn_kvonly_partialrope_decay.py","file_url":"https://github.com/tensorgi/t6/blob/HEAD/model/T6_linearattn_kvonly_partialrope_decay.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":"31eead4c21084b8b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}