{"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/accelerate-parallelizable-reasoning-via","title":"Accelerate Parallelizable Reasoning via Parallel Decoding within One Sequence","arxiv_id":"2503.20533","date":"2025-03-26","proceeding":null,"authors":["Yijiong Yu"],"abstract":"Recent advances in reasoning models have demonstrated significant improvements in accuracy, particularly for complex tasks such as mathematical reasoning, by employing detailed and comprehensive reasoning processes. However, generating these lengthy reasoning sequences is computationally expensive and time-consuming. To address this inefficiency, we leverage the inherent parallelizability of certain tasks to accelerate the reasoning process. Specifically, when multiple parallel reasoning branches exist, we decode multiple tokens per step using a specialized attention mask, processing them within a single sequence. Experimental results show that our method achieves over 100% speedup in decoding time while basically maintaining accuracy.","url_abs":"https://arxiv.org/abs/2503.20533v1","url_pdf":"https://arxiv.org/pdf/2503.20533v1.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":"accelerate-parallelizable-reasoning-via","repo_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"mathematical-reasoning","task_name":"Mathematical Reasoning"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.20533","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.20533"}},"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/yuyijiong/parallel-decoding-in-one-sequence","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/yuyijiong/paralleldecoding-in-one-sequence","reach":{"status":"gone","observed_at":"2026-09-16","how":"tree_404+repo_404"}}],"summary":{"ran_fixture":2,"ran":5,"ran_draft_wrong":2,"unverified":4},"by_repo_kind":{"official":{"samples":10,"ran":6,"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":13,"samples":[{"code_sha256_prefix":"3c76e52815c5401d","entry":"repeat_kv","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":2,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"3c76e52815c5401d"}},{"code_sha256_prefix":"bb3e5ab4d14134fa","entry":"Qwen2Attention","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bb3e5ab4d14134fa"}},{"code_sha256_prefix":"0b7204a99a29e9f0","entry":"Qwen2MLP","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0b7204a99a29e9f0"}},{"code_sha256_prefix":"6641879fc0a39433","entry":"Qwen2PreTrainedModel","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6641879fc0a39433"}},{"code_sha256_prefix":"ea8879988dece177","entry":"Qwen2RMSNorm","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ea8879988dece177"}},{"code_sha256_prefix":"f3c1b1406deb4ad1","entry":"Qwen2RotaryEmbedding","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f3c1b1406deb4ad1"}},{"code_sha256_prefix":"a73c20a30e591545","entry":"_prepare_4d_causal_attention_mask_with_cache_position","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a73c20a30e591545"}},{"code_sha256_prefix":"0fe82a947dc39b42","entry":"apply_rotary_pos_emb","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"0fe82a947dc39b42"}},{"code_sha256_prefix":"e03d53ba9d4f9ae5","entry":"rotate_half","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"e03d53ba9d4f9ae5"}},{"code_sha256_prefix":"3fcec7a2529f7ff3","entry":"Qwen2DecoderLayer","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.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":"3fcec7a2529f7ff3"}},{"code_sha256_prefix":"be6ea9cb7fa07851","entry":"Qwen2FlashAttention2","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.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":"be6ea9cb7fa07851"}},{"code_sha256_prefix":"252125a25cb155ce","entry":"Qwen2ParaModel","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.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":"252125a25cb155ce"}},{"code_sha256_prefix":"c9686309ac575131","entry":"Qwen2SdpaAttention","repo":"yuyijiong/parallel-decoding-in-one-sequence","repo_kind":"official","path":"modeling_qwen2.py","file_url":"https://github.com/yuyijiong/parallel-decoding-in-one-sequence/blob/HEAD/modeling_qwen2.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":"c9686309ac575131"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}