Text-to-Audio (TTA) generation has made rapid progress, but current evaluation methods remain narrow, focusing mainly on perceptual quality while overlooking robustness, generalization, and ethical concerns. We present TTA-Bench, a comprehensive benchmark for evaluating TTA models across functional performance, reliability, and social responsibility. It covers seven dimensions including accuracy, robustness, fairness, and toxicity, and includes 2,999 diverse prompts generated through automated and manual methods. We introduce a unified evaluation protocol that combines objective metrics with over 118,000 human annotations from both experts and general users. Ten state-of-the-art models are benchmarked under this framework, offering detailed insights into their strengths and limitations. TTA-Bench establishes a new standard for holistic evaluation of TTA systems.
论文检索
输入标题、作者或关键词，从 2,854 篇学术成果中精准定位
Large Language Models (LLMs) are increasingly employed in applications that require processing information from heterogeneous formats, including texts, tables, infoboxes, and knowledge graphs. However, systematic biases toward particular formats may undermine LLMs' ability to integrate heterogeneous data impartially, potentially resulting in reasoning errors and increased risks in downstream tasks. Yet it remains unclear whether such biases are systematic, which data-level factors drive them, and what internal mechanisms underlie their emergence. In this paper, we present the first comprehensive study of format bias in LLMs through a three-stage empirical analysis. The first stage explores the presence and direction of bias across a diverse range of LLMs. The second stage examines how key data-level factors influence these biases. The third stage analyzes how format bias emerges within LLMs' attention patterns and evaluates a lightweight intervention to test its effectiveness. Our results show that format bias is consistent across model families, driven by information richness, structure quality, and representation type, and is closely associated with attention imbalance within the LLMs. Based on these investigations, we identify three future research directions to reduce format bias: enhancing data pre-processing through format repair and normalization, introducing inference-time interventions such as attention re-weighting, and developing format-balanced training corpora. These directions will support the design of more robust and fair heterogeneous data processing systems.
Large Language Model (LLM)-based multi-agent systems are increasingly used to simulate human interactions and solve collaborative tasks. A common practice is to assign agents with personas to encourage behavioral diversity. However, this raises a critical yet underexplored question: do personas introduce biases into multi-agent interactions? This paper presents a systematic investigation into persona-induced biases in multi-agent interactions, with a focus on social traits like trustworthiness (how an agent's opinion is received by others) and insistence (how strongly an agent advocates for its opinion). Through a series of controlled experiments in collaborative problem-solving and persuasion tasks, we reveal that (1) LLM-based agents exhibit biases in both trustworthiness and insistence, with personas from historically advantaged groups (e.g., men and White individuals) perceived as less trustworthy and demonstrating less insistence; and (2) agents exhibit significant in-group favoritism, showing a higher tendency to conform to others who share the same persona. These biases persist across various LLMs, group sizes, and numbers of interaction rounds, highlighting an urgent need for awareness and mitigation to ensure the fairness and reliability of multi-agent systems.
Speech-language models (SLMs) offer a promising path toward unifying speech and text understanding and generation. However, challenges remain in achieving effective cross-modal alignment and high-quality speech generation. In this work, we systematically investigate the role of speech tokenizer designs in LLM-centric SLMs, augmented by speech heads and speaker modeling. We compare coupled, semi-decoupled, and fully decoupled speech tokenizers under a fair SLM framework and find that decoupled tokenization significantly improves alignment and synthesis quality. To address the information density mismatch between speech and text, we introduce multi-token prediction (MTP) into SLMs, enabling each hidden state to decode multiple speech tokens. This leads to up to 12× faster decoding and a substantial drop in word error rate (from 6.07 to 3.01). Furthermore, we propose a speaker-aware generation paradigm and introduce RoleTriviaQA, a large-scale role-playing knowledge QA benchmark with diverse speaker identities. Experiments demonstrate that our methods enhance both knowledge understanding and speaker consistency.
Large Language Models (LLMs) have emerged as a new information channel. Meanwhile, one critical but under-explored question is: Is it possible to bypass the safety alignment and inject harmful information into LLMs stealthily? In this paper, we propose to reformulate knowledge editing as a new type of safety threat for LLMs, namely Editing Attack, and conduct a systematic investigation with a newly constructed dataset EditAttack. Specifically, we focus on two typical safety risks of Editing Attack including Misinformation Injection and Bias Injection. For the first risk, we find that editing attacks can inject both commonsense and long-tail misinformation into LLMs, and the effectiveness for the former one is particularly high. For the second risk, we discover that not only can biased sentences be injected into LLMs with high effectiveness, but also one single biased sentence injection can degrade the overall fairness. Then, we further illustrate the high stealthiness of editing attacks. Our discoveries demonstrate the emerging misuse risks of knowledge editing techniques on compromising the safety alignment of LLMs and the feasibility of disseminating misinformation or bias with LLMs as new channels.
Multi-agent reinforcement learning (MARL) is a powerful paradigm for solving cooperative and competitive decision-making problems. While many MARL benchmarks have been proposed, few combine continuous state and action spaces with challenging coordination and planning tasks. We introduce CAMAR, a new MARL benchmark designed explicitly for multi-agent pathfinding in environments with continuous actions. CAMAR supports cooperative and competitive interactions between agents and runs efficiently at up to 100,000 environment steps per second. We also propose a three-tier evaluation protocol to better track algorithmic progress and enable deeper analysis of performance. In addition, CAMAR allows the integration of classical planning methods such as RRT and RRT* into MARL pipelines. We use them as standalone baselines and combine RRT* with popular MARL algorithms to create hybrid approaches. We provide a suite of test scenarios and benchmarking tools to ensure reproducibility and fair comparison. Experiments show that CAMAR presents a challenging and realistic testbed for the MARL community.
Fairness concerns are increasingly critical as machine learning models are deployed in high-stakes applications. While existing fairness-aware methods typically intervene at the model level, they often suffer from high computational costs, limited scalability, and poor generalization. To address these challenges, we propose a Bayesian data selection framework that ensures fairness by aligning group-specific posterior distributions of model parameters and sample weights with a shared central distribution. Our framework supports flexible alignment via various distributional discrepancy measures, including Wasserstein distance, maximum mean discrepancy, and f-divergence, allowing geometry-aware control without imposing explicit fairness constraints. This data-centric approach mitigates group-specific biases in training data and improves fairness in downstream tasks, with theoretical guarantees. Experiments on benchmark datasets show that our method consistently outperforms existing data selection and model-based fairness methods in both fairness and accuracy.
The widespread adoption of graph neural networks (GNNs) has brought increased attention to fairness issues related to sensitive attributes, such as gender and race, in practical scenarios. However, this concern remains largely unexplored in the context of graph clustering. Conventional fair graph clustering methods primarily depend on spectral clustering approaches. Meanwhile, we argue that existing graph learning works mainly focus on a single type of fairness, whereas graph clustering should achieve group equality-informed individual fairness. In this paper, we introduce for the first time a fairness-aware framework termed FairGC for deep graph clustering, which integrates the dual objectives of individual and group fairness while maintaining accurate clustering results. Specifically, we construct two views with distinct semantics using Siamese encoders. Then, we apply multi-step random walks on view-specific affinity graphs to capture high-order affinities of node pairs, thereby reformulating the contrastive learning with a focus on individual similarity. Besides, we utilize adversarial learning by making node representations independent of the estimated sensitive attributes to further eliminate group biases of clustering results. Extensive experiments on four benchmarks demonstrate the effectiveness and superiority of our proposed framework FairGC.
Contribution evaluation is essential for incentivizing high-quality data sharing in federated learning (FL), yet existing Shapley-value-based methods are prohibitively expensive and overlook temporal influence propagation. In this paper, we propose Ripple Shapley, a novel attribution framework that enables accurate, real-time data valuation within a single federated training run. Our method decomposes each sample’s impact into an instantaneous drop term and a recursive ripple term, the latter capturing downstream influence via a Jacobian chain over global updates. To scale computation, we introduce a low-rank approximation of the Jacobian product and construct a shared subspace for efficient ripple accumulation. Extensive experiments on CIFAR-10 and MNIST show that Ripple Shapley achieves up to 62× speedup over existing Shapley-based FL methods while maintaining high attribution fidelity, significantly improving efficiency, robustness, and fairness in federated environments. We further demonstrate its effectiveness in dynamic federated learning scenarios and its potential for real-time data pricing.
Formal verification has emerged as a promising method to ensure the safety and reliability of neural networks. However, many relevant properties, such as fairness or global robustness, pertain to the entire input space. If one applies verification techniques naively, the neural network is checked even on inputs that do not occur in the real world and have no meaning. To tackle this shortcoming, we propose the VeriFlow architecture as a flow-based density model tailored to allow any verification approach to restrict its search to some data distribution of interest. We argue that our architecture is particularly well suited for this purpose because of two major properties. First, we show that the transformation that is defined by our model is piecewise affine. Therefore, the model allows the usage of verifiers based on constraint solving with linear arithmetic. Second, upper density level sets (UDL) of the data distribution are definable via linear constraints in the latent space. As a consequence, representations of UDLs specified by a given probability are effectively computable in the latent space. This property allows for effective verification with a fine-grained, probabilistically interpretable control of how (a-)typical the inputs subject to verification are.
Existing theoretical work on Bayes-optimal fair classifiers usually considers a single (binary) sensitive feature. In practice, individuals are often defined by multiple sensitive features. In this paper, we characterize the Bayes-optimal fair classifier for multiple sensitive features under general approximate fairness measures, including *mean difference* (MD) and *mean ratio* (MR). We show that these approximate measures for existing group fairness notions, including Demographic Parity, Equal Opportunity, Predictive Equality, and Accuracy Parity, are linear transformations of selection rates for specific groups defined by both labels and sensitive features. We then characterize that Bayes-optimal fair classifiers for multiple sensitive features under both MD and MR become instance-dependent thresholding rules that rely on a weighted sum of these group membership probabilities. Our framework applies to both attribute-aware and attribute-blind settings and can accommodate composite fairness notions like Equalized Odds. Building on this, we propose two practical algorithms for Bayes-optimal fair classification via in-processing and post-processing. We show empirically that our methods compare favorably to existing methods.
MULTIBENCH++: A Unified and Comprehensive Multimodal Fusion Benchmarking Across Specialized Domains
PDF ↗Although multimodal fusion has made significant progress, its advancement is severely hindered by the lack of adequate evaluation benchmarks. Current fusion methods are typically evaluated on a small selection of public datasets, a limited scope that inadequately represents the complexity and diversity of real-world scenarios, potentially leading to biased evaluations. This issue presents a twofold challenge. On one hand, models may overfit to the biases of specific datasets, hindering their generalization to broader practical applications. On the other hand, the absence of a unified evaluation standard makes fair and objective comparisons between different fusion methods difficult. Consequently, a truly universal and high-performance fusion model has yet to emerge. To address these challenges, we have developed a large-scale, domain-adaptive benchmark for multimodal evaluation. This benchmark integrates over 30 datasets, encompassing 15 modalities and 20 predictive tasks across key application domains. To complement this, we have also developed an open-source, unified, and automated evaluation pipeline that includes standardized implementations of state-of-the-art models and diverse fusion paradigms. Leveraging this platform, we have conducted large-scale experiments, successfully establishing new performance baselines across multiple tasks. This work provides the academic community with a crucial platform for rigorous and reproducible assessment of multimodal models, aiming to propel the field of multimodal artificial intelligence to new heights.
We propose a multi-agent multi-armed bandit (MA-MAB) framework to ensure fair outcomes across agents while maximizing overall system performance. For example, in a ridesharing setting where a central dispatcher assigns drivers to distinct geographic regions, utilitarian welfare (the sum of driver earnings) can be highly skewed—some drivers may receive no rides. We instead measure fairness by Nash social welfare, i.e., the product of individual rewards. A key challenge in this setting is decision-making under limited information about arm rewards (geographic regions). To address this, we introduce a novel probing mechanism that strategically gathers information about selected arms before assignment. In the offline setting, where reward distributions are known, we exploit submodularity to design a greedy probing algorithm with a constant-factor approximation guarantee. In the online setting, we develop a probing-based algorithm that achieves sublinear regret while preserving Nash social welfare. Extensive experiments on synthetic and real-world datasets demonstrate that our approach outperforms baseline methods in both fairness and efficiency.
With the increasing application of high-stakes decisionmaking application in Federated Learning (FL), ensuring fairness across different populations to prevent biases against certain groups has become crucial. However, achieving group fairness (GF) in FL presents a formidable challenge due to its decentralization, which complicates the global GF estimation by the server. Moreover, distrust and fragility hinder the server from gathering GF values from unreliable clients. This challenge motivates our proposal of OursFed, a provable GF-aware FL framework that integrates a privacy pairbased contract and robust GF estimation method to address issues of distrust and fragility. Methodologically, we categorize client unreliability into two categories: active unreliability stemming from distrust and passive unreliability arising from fragility. To mitigate active unreliability, we design a privacy pair-based contract to guarantee truthful GF reporting, and enhance multivariate analysis by identifying relationships among multiple private data. To counteract passive unreliability, we develop a robust GF estimation using non-parametric techniques to smooth data and estimate probability densities and regression functions, improving per-client GF accuracy under multi-dimensional data perturbation. Theoretically, we demonstrate the efficacy of OursFed by analyzing its convergence, GF stability, and accuracy deviation. Experimentally, evaluations on two real datasets show that OursFed improves GF by 28.61% with at most 2.7% trade-off versus state-ofthe-art baselines, and synthetic experiments further confirm its effectiveness in handling fragility and distrust.
To protect clients' right to be forgotten in federated learning, federated unlearning aims to remove the data contribution of leaving clients from the global learned model. While current studies mainly focused on enhancing unlearning efficiency and effectiveness, the crucial aspects of efficiency fairness and performance fairness among decentralized clients during unlearning have remained largely unexplored. In this study, we introduce FedShard, the first federated unlearning algorithm designed to concurrently guarantee both efficiency fairness and performance fairness. FedShard adaptively addresses the challenges introduced by dilemmas among convergence, unlearning efficiency, and unlearning fairness. Furthermore, we propose two novel metrics to quantitatively assess the fairness of unlearning algorithms, which we prove to satisfy well-known properties in other existing fairness measurements. Our theoretical analysis and numerical evaluation validate FedShard's fairness in terms of both unlearning performance and efficiency. We demonstrate that FedShard mitigates unfairness risks such as cascaded leaving and poisoning attacks and realizes more balanced unlearning costs among clients. Experimental results indicate that FedShard accelerates the data unlearning process 1.3-6.2 times faster than retraining from scratch and 4.9 times faster than the state-of-the-art exact unlearning methods.
Fair clustering is crucial for mitigating bias in unsupervised learning, yet existing algorithms often suffer from quadratic or super-quadratic computational complexity, rendering them impractical for large-scale datasets. To bridge this gap, we introduce the Anchor-based Fair Clustering Framework (AFCF), a novel, general, and plug-and-play framework that empowers arbitrary fair clustering algorithms with linear-time scalability. Our approach first selects a small but representative set of anchors using a novel fair sampling strategy. Then, any off-the-shelf fair clustering algorithm can be applied to this small anchor set. The core of our framework lies in a novel anchor graph construction module, where we formulate an optimization problem to propagate labels while preserving fairness. This is achieved through a carefully designed group-label joint constraint, which we prove theoretically ensures that the fairness of the final clustering on the entire dataset matches that of the anchor clustering. We solve this optimization efficiently using an ADMM-based algorithm. Extensive experiments on multiple large-scale benchmarks demonstrate that AFCF drastically accelerates state-of-the-art methods, which reduces computational time by orders of magnitude while maintaining strong clustering performance and fairness guarantees.
MAU-GPT: Enhancing Multi-type Industrial Anomaly Understanding via Anomaly-aware and Generalist Experts Adaptation
PDF ↗As industrial manufacturing scales, automating fine-grained product image analysis has become critical for quality control. However, existing approaches are hindered by limited dataset coverage and poor model generalization across diverse and complex anomaly patterns. To address these challenges, we introduce MAU-Set, a comprehensive dataset for Multi-type industrial Anomaly Understanding. It spans multiple industrial domains and features a hierarchical task structure, ranging from binary classification to complex reasoning. Alongside this dataset, we establish a rigorous evaluation protocol to facilitate fair and comprehensive model assessment. Building upon this foundation, we further present MAU-GPT, a domain-adapted multimodal large model specifically designed for industrial anomaly understanding. It incorporates a novel AMoE-LoRA mechanism that unifies anomaly-aware and generalist experts adaptation, enhancing both understanding and reasoning across diverse defect classes. Extensive experiments show that MAU-GPT consistently outperforms prior state-of-the-art methods across all domains, demonstrating strong potential for scalable and automated industrial inspection.
Linear models are widely used in high-stakes decision-making due to their simplicity and interpretability. Yet when fairness constraints such as demographic parity are introduced, their effects on model coefficients, and thus on how predictive bias is distributed across features, remain opaque. Existing approaches on linear models often rely on strong and unrealistic assumptions, or overlook the explicit role of the sensitive attribute, limiting their practical utility for fairness assessment. We propose a post-processing framework that can be applied on top of any linear model to decompose the resulting bias into direct (sensitive-attribute) and indirect (correlated-features) components. Our method analytically characterizes how demographic parity reshapes each model coefficient, including those of both sensitive and non-sensitive features. This enables a transparent, feature-level interpretation of fairness interventions and reveals how bias may persist or shift through correlated variables. Our framework requires no model retraining and provides actionable insights for model auditing and mitigation. Experiments on both synthetic and real-world datasets demonstrate that our method captures fairness dynamics missed by prior work, offering a practical and interpretable tool for responsible deployment of linear models.
Fairness-aware machine learning aims to build predictive models that comply with fairness requirements, particularly concerning sensitive attributes such as race, gender, and age. Among causality-based fairness notions, counterfactual fairness is widely adopted for its individual-level guarantees, requiring that an individual’s predicted outcome remains unchanged in a counterfactual world where its sensitive attribute is altered. However, existing methods critically assume that the true causal graph is fully known, which is rarely the case in practice. Moreover, counterfactual fairness suffers from inherent identifiability limitations, as counterfactual quantities cannot always be uniquely estimated from observational data, especially under incomplete causal knowledge. To address these challenges, we propose a principled framework (CF-ICG) for counterfactual fairness under imperfectly known causal graphs, e.g., Completed Partially Directed Acyclic Graphs (CPDAGs). We first introduce a criterion to determine the identifiability, and bound the counterfactual quantities under CPDAGs. Building upon this, we develop an efficient local algorithm that avoids the exhaustive enumeration of all DAGs, ensuring robustness against worst-case fairness violations. Experimental results on synthetic and real-world datasets demonstrate the practical effectiveness and theoretical soundness of CF-ICG.
Graph Neural Networks (GNNs) are expressive architectures for learning from complex graph-structured data. However, their practical use is often limited by the high computational cost of neighborhood aggregation. Recent efforts have focused on knowledge distillation from GNNs to inference-efficient Multi-Layer Perceptrons (MLPs). However, most existing works treat this distillation as an embedding alignment problem, overlooking the need to replicate the topology-aware smoothing behavior that arises from message passing in GNNs. Moreover, existing methods are primarily performance driven, ignoring critical real-world requirements such as fairness. In this work, we make two key observations: (1) state-of-the-art distillation methods fail to capture the heterogeneous smoothness patterns of GNNs, limiting structural awareness in MLPs, and (2) they introduce significant individual and group fairness violations. We introduce FAITH, the first fair and structurally aware GNN-to-MLP distillation framework with graph-free inference. To improve structural awareness in MLPs, we propose a neighborhood-guided energy alignment objective that transfers not only node-level energy, but also the distribution of energies across local neighborhoods. To improve individual fairness, FAITH introduces a novel ℓ2,1-norm objective that preserves structured similarity in the learned representations. Additionally, we incorporate a counterfactual invariance objective that explicitly encourages the model to learn representations that are statistically independent of the sensitive attribute. We provide a comprehensive theoretical analysis of FAITH, interpreting it through a novel instantiation of the Information Bottleneck principle. Extensive experiments on 11 benchmark datasets show that FAITH achieves stronger structural awareness and delivers a better trade-off between utility and fairness than existing methods.