Haiquan Zhao

Lu Lu, Haiquan Zhao, Kan Li et al.

To overcome the performance degradation of adaptive filtering algorithms in the presence of impulsive noise, a novel normalized sign algorithm (NSA) based on a convex combination strategy, called NSA-NSA, is proposed in this paper. The proposed algorithm is capable of solving the conflicting requirement of fast convergence rate and low steady-state error for an individual NSA filter. To further improve the robustness to impulsive noises, a mixing parameter updating formula based on a sign cost function is derived. Moreover, a tracking weight transfer scheme of coefficients from a fast NSA filter to a slow NSA filter is proposed to speed up the convergence rate. The convergence behavior and performance of the new algorithm are verified by theoretical analysis and simulation studies.

Yi Yu, Haiquan Zhao

Proportionate-type normalized suband adaptive filter (PNSAF-type) algorithms are very attractive choices for echo cancellation. To further obtain both fast convergence rate and low steady-state error, in this paper, a variable step size (VSS) version of the presented improved PNSAF (IPNSAF) algorithm is proposed by minimizing the square of the noise-free a posterior subband error signals. A noniterative shrinkage method is used to recover the noise-free a priori subband error signals from the noisy subband error signals. Significantly, the proposed VSS strategy can be applied to any other PNSAF-type algorithm, since it is independent of the proportionate principles. Simulation results in the context of acoustic echo cancellation have demonstrated the effectiveness of the proposed method.

Lu Lu, Haiquan Zhao, Badong Chen

Although the kernel robust mixed-norm (KRMN) algorithm outperforms the kernel least mean square (KLMS) algorithm in impulsive noise, it still has two major problems as follows: (1) The choice of the mixing parameter in the KRMN is crucial to obtain satisfactory performance. (2) The structure of the KRMN algorithm grows linearly as the iteration goes on, thus it has high computational complexity and memory requirements. To solve the parameter selection problem, two variable-mixing parameter KRMN (VPKRMN) algorithms are developed in this paper. Moreover, a sparsification algorithm, quantized VPKRMN (QVPKRMN) algorithm is introduced for nonlinear system identification with impulsive interferences. The energy conservation relation (ECR) and convergence property of the QVPKRMN algorithm are analyzed. Simulation results in the context of nonlinear system identification under impulsive interference demonstrate the superior performance of the proposed VPKRMN and QVPKRMN algorithms as compared with the existing algorithms.

Kaili Yin, Haiquan Zhao, Lu Lu

The bias-compensated set-membership normalised LMS (BCSMNLMS) algorithm is proposed based on the concept of set-membership filtering, which incorporates the bias-compensation technique to mitigate the negative effect of noisy inputs. Moreover, an efficient regression noise variance estimation method is developed by taking the iterative-shrinkage method. Simulations in the context of system identification demonstrate that the misalignment of the proposed BCSM-NLMS algorithm is low for noisy inputs.

Lu Lu, Haiquan Zhao

In this Letter, we present a novel class of diffusion algorithms that can be used to estimate the coefficients of sparse Volterra network (SVN). The development of the algorithms is based on the logarithmic cost and l0-norm constraint. Simulations for Gaussian and impulsive scenarios are conducted to demonstrate the superior performance of the proposed algorithms as compared with the existing algorithms.

Lu Lu, Haiquan Zhao

As a well-established adaptation criterion, the maximum correntropy criterion (MCC) has been receiving increasing attention due to its robust against outliers. In this paper, a new complex recursive maximum correntropy (CRMC) algorithm without any priori information on the noise characteristics, is proposed under the MCC. The proposed algorithm is useful for adaptive beamforming, when the desired signal is contaminated by the impulsive noises. Moreover, the analysis of convergence property of the CRMC algorithm is performed. The results obtained from simulation study establish the effectiveness of this new beamformer.

Lingyu Li, Yixu Wang, Haiquan Zhao et al.

With large language models (LLMs) increasingly deployed as cognitive engines for AI agents, the reliability and effectiveness critically hinge on their intrinsic epistemic agency, which remains understudied. Epistemic agency, the ability to flexibly construct, adapt, and monitor beliefs about dynamic environments, represents a base-model-level capacity independent of specific tools, modules, or applications. We characterize the holistic process underlying epistemic agency, which unfolds in seven interrelated dimensions: prediction, decision-making, perception, memory, counterfactual thinking, belief updating, and meta-reflection. Correspondingly, we propose Reflection-Bench, a cognitive-psychology-inspired benchmark consisting of seven tasks with long-term relevance and minimization of data leakage. Through a comprehensive evaluation of 16 models using three prompting strategies, we identify a clear three-tier performance hierarchy and significant limitations of current LLMs, particularly in meta-reflection capabilities. While state-of-the-art LLMs demonstrate rudimentary signs of epistemic agency, our findings suggest several promising research directions, including enhancing core cognitive functions, improving cross-functional coordination, and developing adaptive processing mechanisms. Our code and data are available at https://github.com/AI45Lab/ReflectionBench.

Jinyi Han, Ying Huang, Ying Liao et al.

Large Reasoning Models (LRMs) have achieved impressive performance on challenging tasks, yet their deep reasoning often incurs substantial computational costs. To achieve efficient reasoning, existing reinforcement learning methods still struggle to construct short reasoning path during the rollout stage, limiting effective learning. Inspired by Evidence Accumulation Models, we find that LRMs have accumulated sufficient information early in reasoning, making further reasoning steps redundant. Based on this insight, we propose Just-Enough Thinking (JET), which trains models to proactively terminate unnecessary reasoning. JET performs trajectory truncation during rollout to expose the model to short, distributionally consistent reasoning paths. Besides, it uses a quality-controlled length reward to better encourage concise reasoning while maintaining correctness. Extensive experiments demonstrate that JET significantly improves reasoning efficiency without sacrificing accuracy. Especially, DeepSeek-Distill-Qwen-1.5B achieves a 4.6% accuracy gain while reducing output length by 46.3% on the Olympiad benchmark. Our code is available in the GitHub.

Jinyi Han, Xinyi Wang, Haiquan Zhao et al.

Recent advances in self-refinement have demonstrated significant potential for improving the outputs of large language models (LLMs) through iterative refinement. However, most existing self-refinement methods rely on a reactive process with a fixed number of iterations, making it difficult to determine the optimal timing and content of refinement based on the evolving generation context. Inspired by the way humans dynamically refine their thoughts during execution, we propose ProActive Self-Refinement (PASR), a novel method that enables LLMs to refine their outputs during the generation process. Unlike methods that regenerate entire responses, PASR proactively decides whether, when, and how to refine based on the model's internal state and evolving context. We conduct extensive experiments on a diverse set of 10 tasks to evaluate the effectiveness of PASR. Experimental results show that PASR significantly enhances problem-solving performance. In particular, on Qwen3-8B, PASR reduces average token consumption by 41.6% compared to standard generation, while also achieving an 8.2% improvement in accuracy. Our code and baselines used in the paper are available on GitHub.

Haiquan Zhao, Lingyu Li, Shisong Chen et al.

Emotion Support Conversation (ESC) is a crucial application, which aims to reduce human stress, offer emotional guidance, and ultimately enhance human mental and physical well-being. With the advancement of Large Language Models (LLMs), many researchers have employed LLMs as the ESC models. However, the evaluation of these LLM-based ESCs remains uncertain. Inspired by the awesome development of role-playing agents, we propose an ESC Evaluation framework (ESC-Eval), which uses a role-playing agent to interact with ESC models, followed by a manual evaluation of the interactive dialogues. In detail, we first re-organize 2,801 role-playing cards from seven existing datasets to define the roles of the role-playing agent. Second, we train a specific role-playing model called ESC-Role which behaves more like a confused person than GPT-4. Third, through ESC-Role and organized role cards, we systematically conduct experiments using 14 LLMs as the ESC models, including general AI-assistant LLMs (ChatGPT) and ESC-oriented LLMs (ExTES-Llama). We conduct comprehensive human annotations on interactive multi-turn dialogues of different ESC models. The results show that ESC-oriented LLMs exhibit superior ESC abilities compared to general AI-assistant LLMs, but there is still a gap behind human performance. Moreover, to automate the scoring process for future ESC models, we developed ESC-RANK, which trained on the annotated data, achieving a scoring performance surpassing 35 points of GPT-4. Our data and code are available at https://github.com/AIFlames/Esc-Eval.

Tianle Gu, Zeyang Zhou, Kexin Huang et al.

Powered by remarkable advancements in Large Language Models (LLMs), Multimodal Large Language Models (MLLMs) demonstrate impressive capabilities in manifold tasks. However, the practical application scenarios of MLLMs are intricate, exposing them to potential malicious instructions and thereby posing safety risks. While current benchmarks do incorporate certain safety considerations, they often lack comprehensive coverage and fail to exhibit the necessary rigor and robustness. For instance, the common practice of employing GPT-4V as both the evaluator and a model to be evaluated lacks credibility, as it tends to exhibit a bias toward its own responses. In this paper, we present MLLMGuard, a multidimensional safety evaluation suite for MLLMs, including a bilingual image-text evaluation dataset, inference utilities, and a lightweight evaluator. MLLMGuard's assessment comprehensively covers two languages (English and Chinese) and five important safety dimensions (Privacy, Bias, Toxicity, Truthfulness, and Legality), each with corresponding rich subtasks. Focusing on these dimensions, our evaluation dataset is primarily sourced from platforms such as social media, and it integrates text-based and image-based red teaming techniques with meticulous annotation by human experts. This can prevent inaccurate evaluation caused by data leakage when using open-source datasets and ensures the quality and challenging nature of our benchmark. Additionally, a fully automated lightweight evaluator termed GuardRank is developed, which achieves significantly higher evaluation accuracy than GPT-4. Our evaluation results across 13 advanced models indicate that MLLMs still have a substantial journey ahead before they can be considered safe and responsible.

Haiquan Zhao, Chenhan Yuan, Fei Huang et al.

As large language models (LLMs) become more capable and widely used, ensuring the safety of their outputs is increasingly critical. Existing guardrail models, though useful in static evaluation settings, face two major limitations in real-world applications: (1) they typically output only binary "safe/unsafe" labels, which can be interpreted inconsistently across diverse safety policies, rendering them incapable of accommodating varying safety tolerances across domains; and (2) they require complete model outputs before performing safety checks, making them fundamentally incompatible with streaming LLM inference, thereby preventing timely intervention during generation and increasing exposure to harmful partial outputs. To address these challenges, we present Qwen3Guard, a series of multilingual safety guardrail models with two specialized variants: Generative Qwen3Guard, which casts safety classification as an instruction-following task to enable fine-grained tri-class judgments (safe, controversial, unsafe); and Stream Qwen3Guard, which introduces a token-level classification head for real-time safety monitoring during incremental text generation. Both variants are available in three sizes (0.6B, 4B, and 8B parameters) and support up to 119 languages and dialects, providing comprehensive, scalable, and low-latency safety moderation for global LLM deployments. Evaluated across English, Chinese, and multilingual benchmarks, Qwen3Guard achieves state-of-the-art performance in both prompt and response safety classification. All models are released under the Apache 2.0 license for public use.

Haiquan Zhao, Xuwu Wang, Shisong Chen et al.

In recent years, multi-modal entity linking (MEL) has garnered increasing attention in the research community due to its significance in numerous multi-modal applications. Video, as a popular means of information transmission, has become prevalent in people's daily lives. However, most existing MEL methods primarily focus on linking textual and visual mentions or offline videos's mentions to entities in multi-modal knowledge bases, with limited efforts devoted to linking mentions within online video content. In this paper, we propose a task called Online Video Entity Linking OVEL, aiming to establish connections between mentions in online videos and a knowledge base with high accuracy and timeliness. To facilitate the research works of OVEL, we specifically concentrate on live delivery scenarios and construct a live delivery entity linking dataset called LIVE. Besides, we propose an evaluation metric that considers timelessness, robustness, and accuracy. Furthermore, to effectively handle OVEL task, we leverage a memory block managed by a Large Language Model and retrieve entity candidates from the knowledge base to augment LLM performance on memory management. The experimental results prove the effectiveness and efficiency of our method.

Lu Lu, Haiquan Zhao, Badong Chen

In this paper, a kernel least mean absolute third (KLMAT) algorithm is developed for adaptive prediction. Combining the benefits of the kernel method and the least mean absolute third (LMAT) algorithm, the proposed KLMAT algorithm performs robustly against noise with different probability densities. To further enhance the convergence rate of the KLMAT algorithm, a variable step-size version (VSS-KLMAT algorithm) is proposed based on a Lorentzian function. Moreover, the stability and convergence property of the proposed algorithms are analyzed. Simulation results in the context of time series prediction demonstrate that the effectiveness of proposed algorithms.

Lu Lu, Haiquan Zhao

The diffusion least-mean square (dLMS) algorithms have attracted much attention owing to its robustness for distributed estimation problems. However, the performance of such filters may change when they are implemented for suppressing noises from speech signals. To overcome this problem, a diffusion leaky dLMS algorithm is proposed in this work, which is characterized by its numerical stability and small misadjustment for noisy speech signals when the unknown system is a lowpass filter. Finally, two implementations of the leaky dLMS are introduced. It is demonstrated that the leaky dLMS can be effectively introduced into a noise reduction network for speech signals.

Yi Yu, Haiquan Zhao, Badong Chen

In order to improve the performances of recently-presented improved normalized subband adaptive filter (INSAF) and proportionate INSAF algorithms for highly noisy system, this paper proposes their set-membership versions by exploiting the theory of set-membership filtering. Apart from obtaining smaller steady-state error, the proposed algorithms significantly reduce the overall computational complexity. In addition, to further improve the steady-state performance for the algorithms, their smooth variants are developed by using the smoothed absolute subband output errors to update the step sizes. Simulation results in the context of acoustic echo cancellation have demonstrated the superiority of the proposed algorithms.

Badong Chen, Lei Xing, Haiquan Zhao et al.

The maximum correntropy criterion (MCC) has recently been successfully applied in robust regression, classification and adaptive filtering, where the correntropy is maximized instead of minimizing the well-known mean square error (MSE) to improve the robustness with respect to outliers (or impulsive noises). Considerable efforts have been devoted to develop various robust adaptive algorithms under MCC, but so far little insight has been gained as to how the optimal solution will be affected by outliers. In this work, we study this problem in the context of parameter estimation for a simple linear errors-in-variables (EIV) model where all variables are scalar. Under certain conditions, we derive an upper bound on the absolute value of the estimation error and show that the optimal solution under MCC can be very close to the true value of the unknown parameter even with outliers (whose values can be arbitrarily large) in both input and output variables. Illustrative examples are presented to verify and clarify the theory.

Badong Chen, Lei Xing, Bin Xu et al.

Nonlinear similarity measures defined in kernel space, such as correntropy, can extract higher-order statistics of data and offer potentially significant performance improvement over their linear counterparts especially in non-Gaussian signal processing and machine learning. In this work, we propose a new similarity measure in kernel space, called the kernel risk-sensitive loss (KRSL), and provide some important properties. We apply the KRSL to adaptive filtering and investigate the robustness, and then develop the MKRSL algorithm and analyze the mean square convergence performance. Compared with correntropy, the KRSL can offer a more efficient performance surface, thereby enabling a gradient based method to achieve faster convergence speed and higher accuracy while still maintaining the robustness to outliers. Theoretical analysis results and superior performance of the new algorithm are confirmed by simulation.

Yi Yu, Haiquan Zhao

Recently, the normalized subband adaptive filter (NSAF) algorithm has attracted much attention for handling the colored input signals. Based on the first-order Markov model of the optimal tap-weight vector, this paper provides a convergence analysis of the standard NSAF. Following the analysis, both the step size and the regularization parameter in the NSAF are jointly optimized in such a way that minimizes the mean square deviation. The resulting joint-optimization step size and regularization parameter (JOSR-NSAF) algorithm achieves a good tradeoff between fast convergence rate and low steady-state error. Simulation results in the context of acoustic echo cancellation demonstrate good features of the proposed algorithm.

Badong Chen, Xi Liu, Haiquan Zhao et al.

Traditional Kalman filter (KF) is derived under the well-known minimum mean square error (MMSE) criterion, which is optimal under Gaussian assumption. However, when the signals are non-Gaussian, especially when the system is disturbed by some heavy-tailed impulsive noises, the performance of KF will deteriorate seriously. To improve the robustness of KF against impulsive noises, we propose in this work a new Kalman filter, called the maximum correntropy Kalman filter (MCKF), which adopts the robust maximum correntropy criterion (MCC) as the optimality criterion, instead of using the MMSE. Similar to the traditional KF, the state mean and covariance matrix propagation equations are used to give prior estimations of the state and covariance matrix in MCKF. A novel fixed-point algorithm is then used to update the posterior estimations. A sufficient condition that guarantees the convergence of the fixed-point algorithm is given. Illustration examples are presented to demonstrate the effectiveness and robustness of the new algorithm.

Wentao Ma, Badong Chen, Jiandong Duan et al.

Robust diffusion adaptive estimation algorithms based on the maximum correntropy criterion (MCC), including adaptation to combination MCC and combination to adaptation MCC, are developed to deal with the distributed estimation over network in impulsive (long-tailed) noise environments. The cost functions used in distributed estimation are in general based on the mean square error (MSE) criterion, which is desirable when the measurement noise is Gaussian. In non-Gaussian situations, such as the impulsive-noise case, MCC based methods may achieve much better performance than the MSE methods as they take into account higher order statistics of error distribution. The proposed methods can also outperform the robust diffusion least mean p-power(DLMP) and diffusion minimum error entropy (DMEE) algorithms. The mean and mean square convergence analysis of the new algorithms are also carried out.

Badong Chen, Lei Xing, Haiquan Zhao et al.

As a robust nonlinear similarity measure in kernel space, correntropy has received increasing attention in domains of machine learning and signal processing. In particular, the maximum correntropy criterion (MCC) has recently been successfully applied in robust regression and filtering. The default kernel function in correntropy is the Gaussian kernel, which is, of course, not always the best choice. In this work, we propose a generalized correntropy that adopts the generalized Gaussian density (GGD) function as the kernel (not necessarily a Mercer kernel), and present some important properties. We further propose the generalized maximum correntropy criterion (GMCC), and apply it to adaptive filtering. An adaptive algorithm, called the GMCC algorithm, is derived, and the mean square convergence performance is studied. We show that the proposed algorithm is very stable and can achieve zero probability of divergence (POD). Simulation results confirm the theoretical expectations and demonstrate the desirable performance of the new algorithm.

Yi Yu, Haiquan Zhao

To overcome the tradeoff of the conventional normalized least mean square (NLMS) algorithm between fast convergence rate and low steady-state misalignment, this paper proposes a variable step size (VSS) NLMS algorithm by devising a new strategy to update the step size. In this strategy, the input signal power and the cross-correlation between the input signal and the error signal are used to estimate the true tracking error power, reducing the effect of the system noise on the algorithm performance. Moreover, the steady-state performances of the algorithm are provided for Gaussian white input signal and are verified by simulations. Finally, simulation results in the context of the system identification and acoustic echo cancellation (AEC) have demonstrated that the proposed algorithm has lower steady-state misalignment than other VSS algorithms.