NSFT (Neural Sub-expert Fine-Tuning), a fine-grained PEFT framework that refines MoE adaptation from experts to sub-experts, is proposed and suggests that sub-expert-level adaptation is a more precise and efficient PEFT paradigm for MoE LLMs.
Abstract
As large language models (LLMs) scale rapidly, dense full-parameter adaptation becomes increasingly expensive, motivating sparse and modular architectures such as Mixture-of-Experts (MoE) models. This shift raises a key question for parameter-efficient fine-tuning (PEFT): at what granularity should parameters be selected and updated? Existing PEFT methods such as LoRA operate on predefined weight matrices, while expert-level sparse tuning methods update entire selected experts. However, we observe that activated experts are internally sparse, with only a small fraction of intermediate channels strongly responding to downstream tasks, indicating that expert-level adaptation is still too coarse. We propose NSFT (Neural Sub-expert Fine-Tuning), a fine-grained PEFT framework that refines MoE adaptation from experts to sub-experts. NSFT decomposes each expert along the intermediate dimension into structured channel groups and selects task-relevant sub-experts by combining routing importance with intra-expert activation saliency. To optimize sparse partial updates, NSFT further introduces learning-rate scaling and dynamic gradient scaling to compensate for the reduced effective update magnitude. Experiments on OLMoE and Ling-mini-2.0 across challenging domain-specific tasks and general benchmarks show that NSFT consistently outperforms representative PEFT and expert-level sparse tuning baselines, while using substantially fewer trainable parameters and preserving competitive general capability. These results suggest that sub-expert-level adaptation is a more precise and efficient PEFT paradigm for MoE LLMs.
Parameter-efficient fine-tuning (PEFT) enables large language models (LLMs) to adapt to specialized tasks, but often at the cost of degrading general capabilities acquired during pretraining. Existing approaches primarily mitigate this trade-off through data replay or regularization, relying on additional data or expli...
Zhi-Qian Pang, Zi-Hong Sun, Qi Xie et al.· 0 citations
The proposed ACE (Adapter Consolidation across Experts), which groups redundant experts and replaces their expert-specific adapters with group-shared higher-rank LoRA modules under the same PEFT budget, achieves the highest observed mean accuracy among the parameter-matched PEFT methods on the three backbones with comp...
Ahin Lee, Sehyun Yun, Joonha Park et al.· 0 citations
Mixture-of-Experts (MoE) architectures scale model capacity through sparse computation, routing each token through only a small subset of experts. In this work, we explore whether this sparsity gives rise to emergent intrinsic organization in multimodal MoEs. We find that experts develop strong semantic specialization...
Damiano Marsili, Raphi Kang, Aditya Mehta et al.· 0 citations
Mixture-of-Experts (MoE) architectures have become essential for scaling large language models, with recent work demonstrating the benefits of fine-grained expert designs. Training such models from scratch is expensive, and sparse upcycling from pre-trained dense models is an attractive alternative. However, we identif...
Yuxuan Lou, Kai Yang, Geng Zhang et al.· 0 citations
MetaNet is proposed, a support-set controller that predicts, for each layer, an expert-retention threshold and a bounded routing bias and provides a tunable accuracy-expert-activation trade-off on DeepSeek-MoE-16B-Chat.
Rong-Feng Wang, Shitao Weng, Zhiquan Wang et al.· 0 citations
FedTAR is proposed, a task-aware federated fine-tuning method for MoE-based LLMs that establishes the association between local updates and task preference via routing outputs and preserves expert specialization and mitigates destructive interference among heterogeneous clients.
Ting-Qi Wang, Hong-Yu Ke, Hao-Xin Wang et al.· arXiv.org· 0 citations
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.