Communication-Efficient Mixture-of-Experts

Graph Neural Networks (GNNs) excel on graph-structured data but suffer catastrophic forgetting when classes arrive sequentially. In class-incremental graph learning, predictions are often dominated by recently learned classes, causing severe recency bias. Dynamic expansion can isolate knowledge across sessions, but many methods assume task-incremental settings and require task identities at test time to select branches or heads---information unavailable in class-incremental learning. Learnable routers offer an alternative, yet continual fine-tuning the router on new data can bias routing toward recent components. We propose MAEM, a replay-free, capacity-controlled dynamic mixture-of-experts framework for class-incremental graph learning. MAEM builds representations with Simplified Graph Convolution (SGC) and trains a dedicated expert for each new session, then freezes it to protect prior knowledge. For routing without task IDs or a shared trainable router, each expert uses a masked autoencoder (MAE); test nodes are assigned to the expert with the lowest reconstruction error, reducing routing-level recency bias. When the expert pool reaches a fixed capacity, MAEM merges the least-used experts via Git Re-Basin alignment and further distills them into a fused expert using topology-aware pseudo data synthesized from MAE distributions (DeepInversion-style), without storing or revisiting historical graph data. On four class-incremental graph benchmarks, MAEM mitigates recency bias, outperforms strong continual graph learning baselines, and keeps model capacity bounded under strict replay-free learning.

Deploying Mixture-of-Experts (MoE) models relies heavily on Expert Parallelism, which generates intense inter-GPU communication. Consequently, state-of-the-art inference systems require high-bandwidth, P2P interconnects (e.g., NVLink) in datacenter GPUs to handle massive token routing, making deployment prohibitively expensive. Consumer GPUs offer comparable compute power at significantly lower cost, promising to democratize MoE inference for individuals and enable privacy-preserving local deployments. However, their bandwidth-limited (only weak PCIe bus bandwidth) and host-mediated interconnects (no P2P support) introduce severe communication bottlenecks. We present CoMoE, a communication-efficient MoE inference system that resolves this mismatch through novel host-centric routing. Our key insight is that the unique communication topology provides the opportunity to elevate the host to an active routing hub, which can fundamentally reduce communication volume and eliminate global synchronization-induced stalls. Specifically, for token dispatch, we introduce host-backed token multicast to write shared tokens to the host exactly once, eliminating outbound transmission redundancy. For token combine, we propose a fine-grained, token-level aggregation mechanism using host staging buffers, which replaces rigid global synchronization and mitigates straggler effects. Evaluation on RTX 5090 GPUs shows that CoMoE improves inference throughput by up to 1.46x, approaching the performance of NVLink-capable A800 GPUs at only 23.4% of the hardware cost.

Mixture-of-Experts (MoE) layers replace the feed-forward block of a Transformer with E expert networks, and each token is routed to k of these experts. Under expert parallelism (EP) the experts are distributed across GPUs, and every MoE layer runs all-to-all collectives in the forward and backward passes to dispatch tokens to their experts and then combine the results. On a cluster with 8 AMD Instinct MI300X GPUs per node, these collectives can take 45% of the training step at EP32 with top-2 routing and 60% with top-6 routing. We find that early in pretraining routers have already learned to assign tokens to experts in correlated patterns, both within a layer and across layers. At top-2, 0.8% of the expert pairs in a layer are selected together by 42% of tokens, and the experts a token selects at one layer predict the experts it selects at the next layer. We use these correlations to keep more token--expert assignments on the token's own GPU, which reduces communication across GPUs and across nodes. Correlated expert placement puts experts that are often selected together on the same GPU. Combined with a dispatcher that sends each token to each GPU once, it removes up to 58% of dispatched rows. Token shuffling applies when sequence parallelism shards tokens across the EP group. It moves each token to the GPU predicted to hold its next-layer experts during the reduce-scatter that follows attention. On one node this raises the share of token--expert assignments served on the token's GPU from 12.5% to 59%. In Megatron-LM, across EP degrees from 8 to 64 with top-2 and top-6 routing, the two methods reduce all-to-all time by 1.16-2.63X and end-to-end step time by up to 1.41X. Neither method changes the models' underlying routing decisions or expert parameters.

When training Mixture-of-Experts (MoE) language models with expert parallelism, all-to-all token dispatch and combine collectives can consume a substantial fraction of end-to-end training time. In this work, we study communication-efficient MoE models (CE-MoE), in which we adopt a heterogeneous layer pattern that decouples token-mixing and channel-mixing depth. Compared to conventional models which interleave MoE layers after each token-mixing layer (e.g., attention, Mamba-2), CE-MoE models concentrate expert capacity in a select few routed MoE layers, while maintaining depth by adding additional token-mixing and dense-FFN layers. Across a scaling ladder from 2B to 31.5B total parameters, under matched total and activated parameters, CE-MoE models consistently reduce training cost while matching validation loss and downstream benchmarks with full-MoE baselines. At the 31.5B scale, CE-MoE uses 33.3% fewer GPU-hours while improving average downstream score and inference throughput.

Mixture-of-experts (MoE) models have recently moved beyond routing a fixed number of complete experts. Shared-expert designs preserve reusable knowledge, fine-grained methods vary computation within experts, and dynamic routers adapt the number of active experts. Yet these decisions are usually made independently, overlooking a basic dependency: extracting reusable computation changes both what remains and how much expert capacity the remainder needs. We study this dependency by decomposing sparsely upcycled feed-forward experts into key-value channels. Co-activated experts align at a subset of value positions; removing these positions changes expert preference; and greater shared coverage is associated with lower residual expert demand. These observations lead to one principle: share first, then route what remains. We instantiate it in UniF-MoE, a unified framework for token-adaptive MoE computation. Each expert is partitioned into aligned blocks. A shared-demand score sets the shared block count and pathway weight, key prototypes select the shared content, and the complementary demand determines the residual expert count through cumulative routing mass. A Gram regularizer separates and normalizes router embeddings, promoting diverse routing directions, sparse expert overlap, and a simple routing geometry. Experiments on DomainBed and GLUE show that this unified design improves predictive performance over representative static and dynamic MoEs while reducing activated computation, inference latency, and memory. Code is available at https://github.com/existence0420/UniF-MoE.

Sparsely activated Mixture-of-Experts (MoE) models scale capacity via conditional computation, but distributed inference suffers from cross-GPU expert communication and routing-induced load imbalance. Existing placement methods reduce this cost by co-locating frequently co-activated experts; however, they derive a single deployment plan from globally aggregated routing traces, thereby averaging away the heterogeneous, task-specific co-activation patterns that actually drive communication in multi-task serving. We observe that expert co-activation is strongly task-conditioned: pairs tightly coupled in one task family are often uncorrelated in another, so effective deployment should group experts by task-aware co-activation rather than by a task-agnostic average. Based on this insight, we propose Task-Aware Coactivation Grouping (TACG), a deployment-time framework that uses family-specific dispatch and co-activation traces to derive per-expert task-family preferences, reweights the co-activation graph so that intra-family locality dominates grouping, and assigns each expert to a primary GPU under exact capacity constraints. To keep the static placement robust under online workload skew, we further introduce Generic Expert Shared Replication (GESR), a lightweight companion that identifies generic experts with consistently central co-activation profiles, replicates them across a small set of secondary GPUs, and applies locality- and load-aware selection at serving time. Experiments on three representative open-source MoE models demonstrate that our framework reduces the average communication cost by 31.39% over the baseline, while preserving an average Jain fairness index of 0.9975. This advantage persists even under severe distribution shifts in the inference data, consistently outperforming strong baselines.

Non-stationary time series forecasting is challenged by evolving distribution shifts that static models struggle to capture. While Mixture-of-Experts (MoE) architectures offer a promising paradigm for decoupling complex drift patterns, existing approaches are limited by fixed expert pools and memoryless routing, hampering their ability to adapt to abrupt regime shifts. To address this, we propose Dynamic TMoE, a framework that unifies architectural evolution with temporal continuity during learning phase. By detecting distribution shifts via Maximum Mean Discrepancy (MMD), we dynamically instantiate heterogeneous experts and prune redundant ones to optimize capacity. Additionally, a temporal memory router leverages recurrent states and an anomaly repository to ensure stable, context-aware expert selection without requiring test-time updates. Experiments on nine benchmarks demonstrate state-of-the-art performance, reducing MSE by 10.4% and MAE by 7.8%. Code is available at https://github.com/andone-07/Dynamic-TMoE.

Mixture-of-Experts (MoE) scales language models efficiently through sparse expert activation, and its dynamic variant further reduces computation by adjusting the activated experts in an input-dependent manner. Existing dynamic MoE methods usually rely on pre-training from scratch or task-specific adaptation, leaving the practical conversion of fully trained MoE underexplored. Enabling such adaptation would directly alleviate the inference costs by allowing easy tokens to bypass unnecessary expert during serving. This paper introduces Zero-Expert Self-Distillation Adaptation (ZEDA), a low-cost framework that transforms post-trained static MoE models into efficient dynamic ones. To stabilize this architectural conversion, ZEDA injects parameter-free zero-output experts into each MoE layer and adapts the augmented model through two-stage self-distillation, utilizing the original MoE as a frozen teacher and applying a group-level balancing loss. On Qwen3-30B-A3B and GLM-4.7-Flash across 11 benchmarks spanning math, code, and instruction following, ZEDA eliminates over 50% of expert FLOPs at marginal accuracy loss. It outperforms the strongest dynamic MoE baseline by 6.1 and 4.0 points on the two models, and delivers ~1.20$\times$ end-to-end inference speedup.

Sparse MoE models achieve a good balance between capacity and compute by routing each token to a small subset of experts. However, in most MoE architectures, once a token is routed, the selected experts process it independently and their outputs are combined via a weighted sum. This leaves open whether enabling communication among them could improve performance. While prior work has raised this question, direct interaction among the active routed experts remains underexplored. In this paper, we propose SDG-MoE (Signed Debate Graph Mixture-of-Experts), a novel architecture that adds a lightweight, iterative deliberation step before final aggregation. SDG-MoE introduces three components: (i) two learned interaction matrices over the active experts, a support graph $A^+$ and a critique graph $A^-$, capturing reinforcing and corrective influences; (ii) a signed message-passing step that updates expert representations before aggregation; and (iii) a disagreement-gated Friedkin-Johnsen-style anchoring that controls deliberation strength while preventing expert drift. Together, these enable a structured deliberation process where interaction strength scales with disagreement and specialization is preserved. We also provide a theoretical analysis establishing stability conditions on expert states and showing that deliberation adds only low-order overhead over the active set. In controlled three-seed pretraining experiments, SDG-MoE improves validation perplexity over both an unsigned graph communication baseline and vanilla MoE, outperforming the strongest baseline by 19.8%, and gives the best external perplexity on WikiText-103, C4, and Paloma among the compared systems.

Mixture-of-Experts (MoE) inference requires large-scale token exchange across devices, making dispatch and combine major bottlenecks in both prefill and decode. Beyond network transfer, routing-driven layout transformation, temporary relay, and output restoration can add substantial overhead. Existing MoE communication paths are often buffer-centric, using explicit inter-process relay and reordering buffers around collective transfer. This report presents a relay-buffer-free communication design for MoE inference acceleration on Ascend systems. The design reorganizes dispatch and combine around direct placement into destination expert windows and direct reading from remote expert windows. Built on globally pooled high-bandwidth memory and symmetric-memory allocation, it removes most intermediate relay and reordering buffers while retaining only lightweight control state, including counts, offsets, and synchronization metadata. We instantiate the design as two schedules for the main phases of MoE inference: a prefill schedule with richer planning state for throughput-oriented execution, and a compact decode schedule for latency-sensitive execution. Experiments on Ascend-based MoE workloads show reduced dispatch and combine latency in both settings. At the serving level, the implementation improves time to first token (TTFT), preserves competitive time per output token (TPOT), and enlarges the feasible scheduling space under practical latency constraints. These results indicate that, on platforms with globally addressable device memory, reducing intermediate buffering and output restoration around expert execution is an effective direction for accelerating MoE inference.

Resource-efficient machine learning increasingly uses sparse Mixture-of-Experts (MoE) architectures, where the gate acts as both a learning component and a routing interface controlling computation, communication, and accuracy. Motivated by finite-rate interpretations of MoE gating, we treat the gate as a stochastic channel and use $I(X;T)$ to quantify the routing information available to the selected expert. To make the associated information quantities tractable beyond synthetic examples, we develop a finite-bank MNIST construction using pretrained CNN experts and a discrete, data-dependent selection rule. Since the selected model belongs to a finite candidate set, the algorithmic mutual information $I(S;W)$ admits a closed-form discrete-entropy estimator from the empirical posterior $q(W|S)$. Sweeping a data-dependence parameter $α$, we observe that $\widehat I(S;W)$ monotonically tracks the generalization gap, while the Xu-Raginsky bound exhibits the expected looseness. We also compare with a uniform union-bound baseline and introduce an empirical estimator of $I(X;T)$ together with a Blahut-Arimoto procedure for tracing an accuracy-rate curve over the expert bank. The proposed framework provides a practical tool for analyzing resource-aware MoE inference systems and for interpreting $I(X;T)$ and $D(R_g)$ as design proxies for efficient expert routing.

Most recent state-of-the-art (SOTA) large language models (LLMs) use Mixture-of-Experts (MoE) architectures to scale model capacity without proportional per-token compute, enabling higher-quality outputs at manageable serving costs. However, MoE inference at scale is fundamentally bottlenecked by expert load imbalance and inefficient token routing, especially in multi-node deployments where tokens are not guaranteed to be routed to local experts, resulting in significant inter-node all-to-all communication overhead. To systematically characterize these challenges, we profile SOTA open-source MoE models, including Llama 4 Maverick, DeepSeek V3-671B, and Qwen3-230B-A22B, on various datasets and collected over 100k real expert activation traces. Upon studying the expert activation patterns, we uncover various persistent properties across all the frontier MoE models: variable expert load imbalance, domain-specific expert activation where expert popularity shifts across task families (code, math, chat, general), and a strong correlation between prefill and decode expert activations. Motivated by these findings, we propose workload-aware micro-batch grouping and an expert placement strategy to maximize token locality to the destination expert, thereby reducing inter-node communication. Across models and datasets, these optimizations help reduce all2all communication data up to 20, resulting in lower MoE decode latency and better accelerator utilization.

Mixture-of-Experts (MoE) architectures have become essential for scaling large language models, driving the development of specialized device-initiated communication libraries such as DeepEP, Hybrid-EP, and others. These libraries demonstrate the performance benefits of GPU-initiated RDMA for MoE dispatch and combine operations. This paper presents NCCL EP (Expert Parallelism), a ground-up MoE communication library built entirely on NCCL's Device API. NCCL EP provides unified ncclEpDispatch and ncclEpCombine primitives with both C and Python interfaces, supporting Low-Latency (LL) mode for inference decoding and High-Throughput (HT) mode for training and inference prefill. LL targets small batch sizes (1-128 tokens) using direct all-to-all RDMA+NVLink mesh connectivity with double-buffered communication for overlapping dispatch and combine phases. HT targets large batches (4096+ tokens) using hierarchical communication that aggregates tokens within NVLink domains before inter-node RDMA transmission. Both modes leverage Device API for both intra- and inter-node communications, taking advantage of its topology awareness and optimized GPU-initiated implementation. We evaluate NCCL EP on an H100-based cluster across multi-node configurations, demonstrating competitive LL kernel performance and presenting end-to-end results with vLLM integration. By building MoE communication natively within NCCL, NCCL EP provides a supported path for expert parallelism on current and emerging NVIDIA platforms.

Mixture-of-Experts (MoE) architectures decompose prediction tasks into specialized expert sub-networks selected by a gating mechanism. This letter adopts a communication-theoretic view of MoE gating, modeling the gate as a stochastic channel operating under a finite information rate. Within an information-theoretic learning framework, {we specialize a mutual-information generalization bound and develop a rate-distortion characterization $D(R_g)$ of finite-rate gating, where $R_g:=I(X; T)$, yielding (under a standard empirical rate-distortion optimality condition) $\mathbb{E}[R(W)] \le D(R_g)+δ_m+\sqrt{(2/m)\, I(S; W)}$. }The analysis yields capacity-aware limits for communication-constrained MoE systems, and numerical simulations on synthetic multi-expert models empirically confirm the predicted trade-offs between gating rate, expressivity, and generalization.