Shurong Wang

Yuqi Pan, Jinghao Zhuang, Yupeng Feng et al.

Scaling context length is reshaping large-model development, yet full-attention Transformers suffer from prohibitive computation and inference bottlenecks at long sequences. A key challenge is to design foundation models that maintain performance and long-context efficiency with minimal training overhead. We introduce SpikingBrain2.0 (SpB2.0), a 5B model that advances both architecture and training efficiency of its predecessor. Our contributions are two-fold. (1) Architectural Innovation: We propose Dual-Space Sparse Attention (DSSA), an inter-layer hybrid of Sparse Softmax Attention (MoBA) and Sparse Linear Attention (SSE), achieving an improved performance-efficiency trade-off for long-context modeling. SpB2.0 further supports dual quantization paths: INT8-Spiking coding enables sparse event-driven computation, while FP8 coding accelerates inference on modern GPUs. (2) Enhanced Training Strategy: We develop an optimized Transformer-to-Hybrid (T2H) pipeline with dual conversion paths for LLMs and VLMs using curated open-source data. Empirically, SpB2.0-5B and SpB2.0-VL-5B recover most of the base Transformer (Qwen3-4B) capability with under 7k A100 GPU hours. SpB2.0 achieves a 10.13x TTFT speedup at 4M context and supports over 10M tokens on 8 A100 GPUs under vLLM, where full-attention models exceed memory limits. It also demonstrates strong cross-platform compatibility, enabling FP8 GPU inference (2.52x speedup at 250k) and efficient neuromorphic execution (64.31% sparsity, with 70.6% and 46.5% area and power reduction at 500MHz). Overall, SpikingBrain2.0 provides a practical pathway for lightweight, multimodal, spiking foundation models, highlighting the potential of combining brain-inspired mechanisms with efficient architectures for resource-constrained and edge scenarios.

Bojian Yin, Shurong Wang, Haoyu Tan et al.

Real-world sequential signals, such as audio or video, contain critical information that is often embedded within long periods of silence or noise. While recurrent neural networks (RNNs) are designed to process such data efficiently, they often suffer from ``memory decay'' due to a rigid update schedule: they typically update their internal state at every time step, even when the input is static. This constant activity forces the model to overwrite its own memory and makes it hard for the learning signal to reach back to distant past events. Here we show that we can overcome this limitation using Selective-Update RNNs (suRNNs), a non-linear architecture that learns to preserve its memory when the input is redundant. By using a neuron-level binary switch that only opens for informative events, suRNNs decouple the recurrent updates from the raw sequence length. This mechanism allows the model to maintain an exact, unchanged memory of the past during low-information intervals, creating a direct path for gradients to flow across time. Our experiments on the Long Range Arena, WikiText, and other synthetic benchmarks show that suRNNs match or exceed the accuracy of much more complex models such as Transformers, while remaining significantly more efficient for long-term storage. By allowing each neuron to learn its own update timescale, our approach resolves the mismatch between how long a sequence is and how much information it actually contains. By providing a principled approach to managing temporal information density, this work establishes a new direction for achieving Transformer-level performance within the highly efficient framework of recurrent modeling.

Shurong Wang, Yuqi Pan, Zhuoyang Shen et al.

Associative memory models are content-addressable memory systems fundamental to biological intelligence and are notable for their high interpretability. However, existing models evaluate the quality of retrieval based on proximity, which cannot guarantee that the retrieved pattern has the strongest association with the query, failing correctness. We reframe this problem by proposing that a query is a generative variant of a stored memory pattern, and define a variant distribution to model this subtle context-dependent generative process. Consequently, correct retrieval should return the memory pattern with the maximum a posteriori probability of being the query's origin. This perspective reveals that an ideal similarity measure should approximate the likelihood of each stored pattern generating the query in accordance with variant distribution, which is impossible for fixed and pre-defined similarities used by existing associative memories. To this end, we develop adaptive similarity, a novel mechanism that learns to approximate this insightful but unknown likelihood from samples drawn from context, aiming for correct retrieval. We theoretically prove that our proposed adaptive similarity achieves optimal correct retrieval under three canonical and widely applicable types of variants: noisy, masked, and biased. We integrate this mechanism into a novel adaptive Hopfield network (A-Hop), and empirical results show that it achieves state-of-the-art performance across diverse tasks, including memory retrieval, tabular classification, image classification, and multiple instance learning.

Lyudong Jin, Yanning Zhang, Yanhan Li et al.

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. Exploiting the heterogeneous capabilities of edge LLMs is crucial for diverse emerging applications, as it enables greater cost-effectiveness and reduced latency. In this work, we introduce \textit{Mixture-of-Edge-Experts (MoE$^2$)}, a novel collaborative inference framework for edge LLMs. We formulate the joint gating and expert selection problem to optimize inference performance under energy and latency constraints. Unlike conventional MoE problems, LLM expert selection is significantly more challenging due to the combinatorial nature and the heterogeneity of edge LLMs across various attributes. To this end, we propose a two-level expert selection mechanism through which we uncover an optimality-preserving property of gating parameters across expert selections. This property enables the decomposition of the training and selection processes, significantly reducing complexity. Furthermore, we leverage the objective's monotonicity and design a discrete monotonic optimization algorithm for optimal expert selection. We implement edge servers with NVIDIA Jetson AGX Orins and NVIDIA RTX 4090 GPUs, and perform extensive experiments. Our results validate that performance improvements of various LLM models and show that our MoE$^2$ method can achieve optimal trade-offs among different delay and energy budgets, and outperforms baselines under various system resource constraints.

Shurong Wang, Yufei Zhang, Xuliang Huang et al.

Knowledge graph embedding (KGE) has caught significant interest for its effectiveness in knowledge graph completion (KGC), specifically link prediction (LP), with recent KGE models cracking the LP benchmarks. Despite the rapidly growing literature, insufficient attention has been paid to the cooperation between humans and AI on KG. However, humans' capability to analyze graphs conceptually may further improve the efficacy of KGE models with semantic information. To this effect, we carefully designed a human-AI team (HAIT) system dubbed KG-HAIT, which harnesses the human insights on KG by leveraging fully human-designed ad-hoc dynamic programming (DP) on KG to produce human insightful feature (HIF) vectors that capture the subgraph structural feature and semantic similarities. By integrating HIF vectors into the training of KGE models, notable improvements are observed across various benchmarks and metrics, accompanied by accelerated model convergence. Our results underscore the effectiveness of human-designed DP in the task of LP, emphasizing the pivotal role of collaboration between humans and AI on KG. We open avenues for further exploration and innovation through KG-HAIT, paving the way towards more effective and insightful KG analysis techniques.