Author name cluster

Zehao Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

7 papers

1 author row

AAAI Conference 2026 Conference Paper

GLOBA: Rethinking Parameter Conflicts in Model Merging

Zehao Liu
Kun Li
Wei Zhou

Model merging serves as a training-free technique that combines multiple task-specific models into a unified multi-task model, but parameter conflicts often lead to performance drops. Previous methods flatten weight matrices into one-dimensional vectors, losing the inherent structural information of their row and column spaces. We mathematically prove and experimentally validate that parameter conflicts arise from non-orthogonal components of task vectors, while orthogonal components are conflict-free. Furthermore, we find that non-orthogonal components can contain both harmful conflicts and beneficial synergies. To precisely locate parameter conflicts and extract orthogonal components, we propose GLOBA (GLObal Basis Analysis Framework), which projects task vectors onto a global basis to align them within a unified coordinate system and construct a task interaction matrix. Following energy-based pruning, we divide parameters into five types based on the orthogonal relationships between the row spaces and column spaces of task vectors. Experiments on three fine-tuned models (mathematics, coding, and instruction-following) using LLaMA-2-7B and LLaMA-2-13B demonstrate significant performance gains through selective retention of beneficial parameters and removal of conflicting ones.

PDF Details DOI

AAAI Conference 2026 Conference Paper

Model-Agnostic Sentiment Distribution Stability Analysis for Robust LLM-Generated Texts Detection

Siyuan Li
Xi Lin
Guangyan Li
Zehao Liu
Aodu Wulianghai
Li Ding
Jun Wu
Jianhua Li

The rapid advancement of large language models (LLMs) has resulted in increasingly sophisticated AI-generated content, posing significant challenges in distinguishing LLM-generated text from human-written language. Existing detection methods, primarily based on lexical heuristics or fine-tuned classifiers, often suffer from limited generalizability and are vulnerable to paraphrasing, adversarial perturbations, and cross-domain shifts. In this work, we propose SentiDetect, a model-agnostic framework for detecting LLM-generated text by analyzing the divergence in sentiment distribution stability. Our method is motivated by the empirical observation that LLM outputs tend to exhibit emotionally consistent patterns, whereas human-written texts display greater emotional variability. To capture this phenomenon, we define two complementary metrics: sentiment distribution consistency and sentiment distribution preservation, which quantify stability under sentiment-altering and semantic-preserving transformations. We evaluate SentiDetect on five diverse domains and a range of advanced LLMs, including Gemini-1.5-Pro, Claude-3, GPT-4-0613, and LLaMa-3.3. Experimental results demonstrate its superiority over state-of-the-art baselines, with over 16% and 11% F1 score improvements on Gemini-1.5-Pro and GPT-4-0613, respectively. Moreover, SentiDetect also shows greater robustness to paraphrasing, adversarial attacks, and text length variations, outperforming existing detectors in challenging scenarios.

PDF Details DOI

TMLR Journal 2026 Journal Article

SpikingBrain: Spiking Brain-inspired Large Models

Yuqi Pan
Yupeng Feng
JingHao Zhuang
siyu ding
Han Xu
Zehao Liu
Bohan Sun
Yuhong Chou

Mainstream Transformer-based large language models (LLMs) face significant efficiency bottlenecks: training computation scales quadratically with sequence length, and inference memory grows linearly. These constraints limit their ability to process long sequences effectively. In addition, building large models on non-NVIDIA computing platforms poses major challenges in achieving stable and efficient training and deployment. To address these issues, we introduce SpikingBrain, a new family of brain-inspired models designed for efficient long-context training and inference. SpikingBrain leverages the MetaX GPU cluster and focuses on three core aspects: (1) Model Architecture: linear and hybrid-linear attention architectures with adaptive spiking neurons; (2) Algorithmic Optimizations: an efficient, conversion-based training pipeline compatible with existing LLMs, along with a dedicated spike coding framework; (3) System Engineering: customized training frameworks, operator libraries, and parallelism strategies tailored to the MetaX hardware. Using these techniques, we develop two models: SpikingBrain-7B, a linear LLM, and SpikingBrain-76B, a hybrid-linear MoE LLM. These models demonstrate the feasibility of large-scale LLM development on non-NVIDIA platforms, and our training framework supports weeks of stable training on hundreds of MetaX GPUs with Model FLOPs Utilization (MFU) at expected levels. SpikingBrain achieves performance comparable to open-source Transformer baselines while using exceptionally low data resources (continual pre-training of approximately 150B tokens). Our models also significantly improve long-context efficiency and deliver inference with (partially) constant memory and event-driven spiking behavior. For example, SpikingBrain-7B achieves more than 100× speedup in Time to First Token (TTFT) for 4M-token sequences. Furthermore, the proposed spiking scheme achieves 69.15% sparsity, enabling low-power operation. Overall, this work demonstrates the potential of brain-inspired mechanisms to drive the next generation of efficient and scalable large model design.

PDF Details

JBHI Journal 2025 Journal Article

Continuous Estimation of FES-Induced Neuromuscular Fatigue Using Mechanomyography Signals

Zehao Liu
Weiguang Huo
Zhenhua Yu
Paul Bentley
Anthony M. J. Bull
Ravi Vaidyanathan

Functional Electrical Stimulation (FES), a key therapy for improving extremity function (e. g. , in post-stroke patients), is limited by rapid FES-induced muscle fatigue. Additionally, Electromyography (EMG) monitoring is significantly compromised by FES artifacts. Mechanomyography (MMG), directly immune to such electrical FES artifacts, offers a promising alternative for fatigue estimation; however, its quantitative use for closed-loop FES remains underdeveloped. This study validated an MMG-based FES fatigue assessment system, introducing a novel wearable sensor (pressure P_MMG, microphone M_MMG) and an MMG-driven Tibialis Anterior (TA) musculotendon model with an MMG-derived fatigue index. An isometric FES fatigue protocol was conducted on control ( $N=15$ ) and post-stroke ( $N=3$ ) participants, recording force and MMG signals. P_MMG Mean Value (MV) signals consistently decreased with fatigue, showing strong average Pearson correlations ( $\bar{r}$ ) with force decline in both control ( $\bar{r}=0. 740$ ) and stroke ( $\bar{r}=0. 928$ ) groups ( $p \leq 0. 005$ ). Conversely, M_MMG signals exhibited inconsistent trends and weaker force correlations, largely due to non-monotonic behavior in many participants. The P_MMG MV-driven model accurately predicted force decline, achieving mean coefficients of determination ( $R^{2}$ ) of 0. 741 (control) and 0. 774 (stroke), with strong prediction correlations ( $\bar{r} > 0. 87, p < 0. 01$ ). Model predictions utilizing M_MMG signals were successful only for participant subsets with consistent signal trends. The pressure-based P_MMG sensor provided a robust, non-invasive FES-induced fatigue indicator. The P_MMG-driven model allows continuous estimation of force capacity decline, promising for closed-loop FES to optimize rehabilitation.

Details DOI

AAAI Conference 2025 Conference Paper

GCAD: Anomaly Detection in Multivariate Time Series from the Perspective of Granger Causality

Zehao Liu
Mengzhou Gao
Pengfei Jiao

Multivariate time series anomaly detection has numerous real-world applications and is being extensively studied. Modeling pairwise correlations between variables is crucial. Existing methods employ learnable graph structures and graph neural networks to explicitly model the spatial dependencies between variables. However, these methods are primarily based on prediction or reconstruction tasks, which can only learn similarity relationships between sequence embeddings and lack interpretability in how graph structures affect time series evolution. In this paper, we designed a framework that models spatial dependencies using interpretable causal relationships and detects anomalies through changes in causal patterns. Specifically, we propose a method to dynamically discover Granger causality using gradients in nonlinear deep predictors and employ a simple sparsification strategy to obtain a Granger causality graph, detecting anomalies from a causal perspective. Experiments on real-world datasets demonstrate that the proposed model achieves more accurate anomaly detection compared to baseline methods.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

ZeCO: Zero-Communication Overhead Sequence Parallelism for Linear Attention

Yuhong Chou
Zehao Liu
Rui-jie Zhu
Xinyi Wan
Tianjian Li
Congying Chu
Qian Liu
Jibin Wu

Linear attention mechanisms deliver significant advantages for Large Language Models (LLMs) by providing linear computational complexity, enabling efficient processing of ultra-long sequences (e. g. , 1M context). However, existing Sequence Parallelism (SP) methods, essential for distributing these workloads across devices, become the primary performance bottleneck due to substantial communication overhead. In this paper, we introduce ZeCO (Zero Communication Overhead) sequence parallelism for linear attention models, a new SP method designed to overcome these limitations and achieve practically end-to-end near-linear scalability for long sequence training. For example, training a model with a 1M sequence length across 64 devices using ZeCO takes roughly the same time as training with an 16k sequence on a single device. At the heart of ZeCO lies All-Scan, a novel collective communication primitive. All-Scan provides each SP rank with precisely the initial operator state it requires while maintaining a minimal communication footprint, effectively eliminating communication overhead. Theoretically, we prove the optimaity of ZeCO, showing that it introduces only negligible time and space overhead. Empirically, we compare the communication costs of different sequence parallelism strategies and demonstrate that All-Scan achieves the fastest communication in SP scenarios. Specifically, on 256 GPUs with an 8M sequence length, ZeCO achieves a 60\% speedup compared to the current state-of-the-art (SOTA) SP method. We believe ZeCO establishes a clear path toward efficiently training next-generation LLMs on previously intractable sequence lengths.

PDF Details

TIST Journal 2022 Journal Article

Self-supervised Discriminative Representation Learning by Fuzzy Autoencoder

Wenlu Yang
Hongjun Wang
Yinghui Zhang
Zehao Liu
Tianrui Li

Representation learning based on autoencoders has received great concern for its potential ability to capture valuable latent information. Conventional autoencoders pursue minimal reconstruction error, but in most machine learning tasks such as classification and clustering, the discrimination of feature representation is also important. To address this limitation, an enhanced self-supervised discriminative fuzzy autoencoder (FAE) is innovatively proposed, which focuses on exploring information within data to guide the unsupervised training process and enhancing feature discrimination in a self-supervised manner. In FAE, fuzzy membership is applied to provide a means of self-supervised, which allows FAE can not only utilize AE’s outstanding representation learning capabilities but can also transform the original data into another space with improved discrimination. First, the objective function corresponding to FAE is proposed by reconstruction loss and clustering oriented loss simultaneously. Subsequently, Mini-Batch Gradient Descent is applied to infer the objective function and the detailed process is illustrated step by step. Finally, empirical studies on clustering tasks have demonstrated the superiority of FAE over the state of the art.

Details DOI