{"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/arxiv-2602-07425","title":"Sign-Based Optimizers Are Effective Under Heavy-Tailed Noise","arxiv_id":"2602.07425","date":"2026-02-07","proceeding":null,"authors":["Dingzhi Yu","Hongyi Tao","Yuanyu Wan","Luo Luo","Lijun Zhang"],"abstract":"While adaptive gradient methods are the workhorse of modern machine learning, sign-based optimization algorithms such as Lion and Muon have recently demonstrated superior empirical performance over AdamW in training large language models (LLM). However, a theoretical understanding of why sign-based updates outperform variance-adapted methods remains elusive. In this paper, we aim to bridge the gap between theory and practice through the lens of heavy-tailed gradient noise, a phenomenon frequently observed in language modeling tasks. Theoretically, we introduce a novel generalized heavy-tailed noise condition that captures the behavior of LLMs more accurately than standard finite variance assumptions. Under this noise model, we establish sharp convergence rates of SignSGD and Lion for generalized smooth function classes, matching or surpassing previous best-known bounds. Furthermore, we extend our analysis to Muon and Muonlight, providing what is, to our knowledge, the first rigorous analysis of matrix optimization under heavy-tailed stochasticity. These results offer a strong theoretical justification for the empirical superiority of sign-based optimizers, showcasing that they are naturally suited to handle the noisy gradients associated with heavy tails. Empirically, LLM pretraining experiments validate our theoretical insights and confirm that our proposed noise models are well-aligned with practice.","url_abs":"https://arxiv.org/abs/2602.07425","url_pdf":"https://arxiv.org/pdf/2602.07425","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2602.07425","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2602.07425"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs","reach":{"status":"ok"}}],"summary":{"ran":6,"unverified":2},"by_repo_kind":{"found_in_text":{"samples":8,"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":8,"samples":[{"code_sha256_prefix":"1f1b81b26e97e55b","entry":"entropy_reg","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/models/moe.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/models/moe.py","link_basis":"plan_row","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":"1f1b81b26e97e55b"}},{"code_sha256_prefix":"a69212d334eb9bc7","entry":"load_balancing_loss_","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/models/moe.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/models/moe.py","link_basis":"plan_row","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":"a69212d334eb9bc7"}},{"code_sha256_prefix":"2f520af81f2675ca","entry":"log_mean","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/models/moe.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/models/moe.py","link_basis":"plan_row","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":"2f520af81f2675ca"}},{"code_sha256_prefix":"1bd86a911c3f78a3","entry":"move_to_cpu","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/logger/logger.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/logger/logger.py","link_basis":"plan_row","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":"1bd86a911c3f78a3"}},{"code_sha256_prefix":"a1d6f89d43fc42e1","entry":"precompute_freqs_cis","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/models/llama.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/models/llama.py","link_basis":"plan_row","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":"a1d6f89d43fc42e1"}},{"code_sha256_prefix":"198aea734522a443","entry":"self_preserving_overwrite","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/logger/logger.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/logger/logger.py","link_basis":"plan_row","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":"198aea734522a443"}},{"code_sha256_prefix":"9891174b43bd9c9a","entry":"apply_rotary_emb","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/models/llama.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/models/llama.py","link_basis":"plan_row","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":"9891174b43bd9c9a"}},{"code_sha256_prefix":"2fc6fdb85f13dc41","entry":"none_or_str","repo":"Dingzhen230/Heavy-tailed-Noise-in-LLMs","repo_kind":"found_in_text","path":"src/config/base.py","file_url":"https://github.com/Dingzhen230/Heavy-tailed-Noise-in-LLMs/blob/HEAD/src/config/base.py","link_basis":"plan_row","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":"2fc6fdb85f13dc41"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":{"kind":"leaderboard_placements","source":"Syntology's leaderboard-shaped extractor over the paper's own arXiv-HTML tables: a model pointed at a cell, the number was read from that cell and checked against the board's metric, dataset, split and scale, and an independent check accepted the entry; not reviewed by the paper's authors or the archive's editors","extractor_model":"global.anthropic.claude-sonnet-4-5-20250929-v1:0","verifier_model":null,"prompt_sha":"fa63d4bb9d755694","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","papers_newer_than_archive":9581,"papers_checked":6264},"entries":[],"not_placed":{"boards":0,"rejected_by_independent_check":0,"refused_by_a_rule":0,"check_did_not_answer":0,"proposed_without_a_cell":0,"declined_by_site":0}}}