{"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-2605-14200","title":"How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization","arxiv_id":"2605.14200","date":"2026-05-13","proceeding":null,"authors":["Leena Chennuru Vankadara","Moritz Haas","Luke Hayward","Sebastian Bordt","Alessandro Breccia"],"abstract":"Recent frontier large language models predominantly rely on Mixture-of-Experts (MoE) architectures. Despite empirical progress, there is still no principled understanding of how hyperparameters should scale with network width $N$, expert width $N_e$, number of experts $M$, sparsity $K$, and depth $L$ to ensure both stability and optimal performance at scale. We take a principled step toward resolving this gap by analyzing three different scaling regimes: (I) co-scaling $N\\asymp N_e$, (II) co-scaling $N\\asymp M\\asymp K$, and (III) full proportional scaling of $N, N_e, M$, and $K$. For each regime, we develop a novel Dynamical Mean Field Theory (DMFT) description of the limiting training dynamics of MoEs that provides a formal foundation for our analysis. Within this framework, we derive the unique parameterization for SGD and Adam satisfying all maximal-update ($μ$) desiderata. We then show that the resulting $μ$P prescription does not reliably induce monotonic improvement with scale or robust learning-rate transfer. We trace these pathologies to scale-dependent observables in the aggregation dynamics, which motivates a refined set of desiderata that we term maximal scale stability. Guided by this principle, we derive a Maximally Scale-Stable Parameterization (MSSP) for both SGD and Adam in all three scaling regimes, and characterize the corresponding limiting dynamics - qualitatively distinct from the $μ$P limit - through a separate DMFT analysis. Experiments verify that MSSP robustly recovers learning rate transfer and monotonic improvement with scale across regimes. Combined with existing depth-scaling theory, these results provide a complete scaling prescription for MoE architectures as a function of width, depth, expert width, and number of experts.","url_abs":"https://arxiv.org/abs/2605.14200","url_pdf":"https://arxiv.org/pdf/2605.14200","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=2605.14200","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2605.14200"}},"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/vankadara-lab/mssp-moe","reach":null}],"summary":{"ran":9,"ran_draft_wrong":3,"ran_honours":1,"ran_violates":1,"unverified":11},"by_repo_kind":{"found_in_text":{"samples":25,"ran":14,"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":"79206b63d6c2a4cc","entry":"CausalSelfAttention","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"79206b63d6c2a4cc"}},{"code_sha256_prefix":"eb4439589ee0f401","entry":"Expert","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"eb4439589ee0f401"}},{"code_sha256_prefix":"89edafb2b7dd86d4","entry":"MLP","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"89edafb2b7dd86d4"}},{"code_sha256_prefix":"0701f790e1bfa186","entry":"MLPExperts","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0701f790e1bfa186"}},{"code_sha256_prefix":"d6fd8fe0a21c5705","entry":"MOEManager","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d6fd8fe0a21c5705"}},{"code_sha256_prefix":"c4dfde8cc3a53a1e","entry":"MupBottleneck","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c4dfde8cc3a53a1e"}},{"code_sha256_prefix":"cb99fcd3b1f748d2","entry":"MupDense","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cb99fcd3b1f748d2"}},{"code_sha256_prefix":"b0c26f0b82672510","entry":"MupFixedE","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b0c26f0b82672510"}},{"code_sha256_prefix":"c4751b9b12f74641","entry":"ScalingConfig","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c4751b9b12f74641"}},{"code_sha256_prefix":"5d32ec6879733f85","entry":"apply_rope","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5d32ec6879733f85"}},{"code_sha256_prefix":"60caeaf416ffdf78","entry":"build_rope_cache","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"60caeaf416ffdf78"}},{"code_sha256_prefix":"5bc29c2859687b6c","entry":"classify_parameter","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5bc29c2859687b6c"}},{"code_sha256_prefix":"68a803a8db3489f0","entry":"compute_fans","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"68a803a8db3489f0"}},{"code_sha256_prefix":"4053fe32ebf78105","entry":"is_input_like","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"4053fe32ebf78105"}},{"code_sha256_prefix":"6a082740ae6a704b","entry":"Block","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"6a082740ae6a704b"}},{"code_sha256_prefix":"c21ab39f1c8ed9db","entry":"GPT","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"c21ab39f1c8ed9db"}},{"code_sha256_prefix":"908e9b4c72c5ccfd","entry":"MonitorMoELayer","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"908e9b4c72c5ccfd"}},{"code_sha256_prefix":"137a06276c29cba3","entry":"MupAllScale","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"137a06276c29cba3"}},{"code_sha256_prefix":"731519ad66b94ad5","entry":"MupBaselineAllScale","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"731519ad66b94ad5"}},{"code_sha256_prefix":"e46ca229f04a0481","entry":"MupBottleneckNoExpertDownInit","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"e46ca229f04a0481"}},{"code_sha256_prefix":"0401dd09f1f20363","entry":"NanoMoELayer","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"0401dd09f1f20363"}},{"code_sha256_prefix":"b19d401a089c2b4d","entry":"Router","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"b19d401a089c2b4d"}},{"code_sha256_prefix":"0d725e3597c5b8d0","entry":"SoftMoELayer","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"0d725e3597c5b8d0"}},{"code_sha256_prefix":"03ca8589472f37ff","entry":"StandardScaling","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"03ca8589472f37ff"}},{"code_sha256_prefix":"f9f552a09c0196b1","entry":"get_scaling_config","repo":"vankadara-lab/mssp-moe","repo_kind":"found_in_text","path":"transformer-moe-experiments/model.py","file_url":"https://github.com/vankadara-lab/mssp-moe/blob/HEAD/transformer-moe-experiments/model.py","link_basis":"first_harvest_node","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":"f9f552a09c0196b1"}}]},"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}}}