Class-Incremental Learning (CIL) aims to enable models to sequentially learn new tasks while retaining knowledge from previous ones. Recently, merging-based pre-trained CIL methods have gained significant attention due to their competitive performance and high inference efficiency. However, most existing approaches decouple training from merging, neglecting the compatibility among task-specific adapters. This incompatibility introduces severe conflicts during integration, resulting in catastrophic forgetting and notable performance degradation. To address this limitation, we propose MOBO, a Merging-Oriented Bi-Level Optimization framework that synergizes the optimization of the current task model with the performance of the final merged model. At the upper level, we introduce a global loss that anticipates the merged model's behavior, guiding the current task parameters toward a solution space that facilitates effective merging. At the lower level, we employ a dynamic weighted merging strategy to optimize merging coefficients and update the merged model. By alternately optimizing the task-specific and merged models, MOBO effectively mitigates the performance loss caused by incompatible adapter integration. Comprehensive experiments on CIFAR-100, CUB-200, ImageNet-R, ImageNet-A, and VTAB demonstrate the superiority of our approach, highlighting its robustness and scalability, particularly on long task sequences.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
Detecting text generated by large language models~(LLMs) is crucial but challenging. Existing detectors depend on impractical assumptions, such as white-box settings, or solely rely on text-level features, leading to imprecise detection ability. In this paper, we propose a simple but effective and practical LLM-generated text detection method, VaryBalance. The core of VaryBalance is that, compared to LLM-generated texts, there is a greater difference between human texts and their rewritten version via LLMs. Leveraging this observation, VaryBalance quantifies this through mean standard deviation and distinguishes human texts and LLM-generated texts. Comprehensive experiments demonstrated that VaryBalance outperforms the state-of-the-art detectors, i.e., Binoculars, by up to 34.3% in terms of AUROC, and maintains robustness against multiple generating models and languages.
Context-Aware Multi-Agent Coordination: Learning Correlated Equilibria Under Situational Constraints
Effective multi-agent coordination requires aligning incentives while adhering to complex requirements. However, real-world systems often impose situational constraints, context-dependent requirements triggered only under specific conditions, which challenge standard Correlated Equilibria (CE) solutions. We propose Situational-Constrained Density-Based Correlated Equilibria (SC-DBCE), a novel concept in Markov Games that formalizes situational constraints as logic implications. To solve this, we introduce Situational-Constrained Correlated Policy Iteration (SC-CPI), a reinforcement learning algorithm employing a smooth Log-Sum-Exp mechanism for constraint optimization. Evaluations on multi-agent games, smart grids, and warehouse robotics demonstrate that SC-CPI consistently outperforms baselines in both equilibrium quality and constraint adherence. To our knowledge, this is the first method learning CE under situational constraints.
Social bots threaten online platforms by spreading disinformation and manipulating public discourse. Graph neural networks have emerged as effective tools for bot detection by modeling user interactions, yet two fundamental challenges limit their practical deployment: severe class imbalance where bots constitute a small minority of users, and camouflaged edges where bots forge deceptive connections to humans to evade detection. Class imbalance causes decision boundaries to shift toward the majority class, while camouflaged edges corrupt neighborhood aggregation through spurious message passing. We present BotVA, a unified framework addressing both challenges through variational feature augmentation and adversarial graph learning. Our approach makes three key contributions: (i) a conditional variational autoencoder that models minority class distributions and synthesizes semantically coherent features, effectively expanding minority support in representation space; (ii) an adversarial training paradigm where a generator simulates camouflage by injecting deceptive edges while a graph transformer discriminator with semantic attention learns to identify and downweight such perturbations; and (iii) a two-stage training strategy that pretrains the variational module before alternating generator-discriminator optimization to ensure stable convergence. Experiments on three benchmarks demonstrate that BotVA achieves state-of-the-art accuracy and F1-score, exhibits strong robustness under camouflage perturbations, and maintains competitive performance with limited supervision.
Multimodal semantic communication systems face a critical challenge in extracting and aligning semantic features across heterogeneous modalities within a unified representation space, particularly under noisy transmission conditions. To address this, we propose Bridge, a cross-modal learning framework that integrates video, audio, and text into a unified semantic space through feature disentanglement and contrastive alignment. Bridge separates modality-specific and modality-invariant representations, enhancing both intra-modal precision and inter-modal generalization. During decoding, it leverages heterogeneous foundation models to preserve semantic fidelity under channel noise. Extensive experiments on multiple multimodal benchmarks under varying SNR levels demonstrate that Bridge maintains high semantic consistency and reconstruction quality, improving image reconstruction by approximately 16.3% in low-SNR regimes. Our work provides a practical and theoretically grounded framework for next-generation multimodal semantic communication systems.
Treewidth is a fundamental graph invariant that quantifies how tree-like a given graph is. It is extensively used with dynamic programming to design fixed-parameter tractable algorithms for many NP-hard graph combinatorial optimization problems. However, despite broad theoretical applicability, treewidth dynamic programming (TDP) does not scale in practice beyond graphs with very small treewidth. Rather than applying TDP as a standalone technique, we demonstrate that TDP can serve as a broadly applicable enhancer for a wide range of graph combinatorial optimization algorithms. Our framework leverages the concept of treewidth modulators, which refer to vertex sets whose removal significantly reduces the treewidth. We further propose an empirically efficient procedure for generating such treewidth modulators. To enhance an algorithm 𝒜, we use 𝒜 to heuristically make decisions on the modulators vertices, after which the remaining decisions outside the treewidth modulators become scalable for TDP. To demonstrate the general applicability of our proposed framework. We experimented with three classic graph combinatorial optimization models: Maximum Independent Set, Minimum Vertex Cover, and Max Cut. We apply TDP to enhance algorithms across diverse paradigms, including evolutionary search, greedy heuristics, and graph-neural-network-based heuristics. For all combinations of optimization models and base algorithms, TDP significantly improves performance over the original methods. In many settings, TDP-enhanced greedy heuristics are competitive with, and sometimes clearly outperform, state-of-the-art commercial solvers.
Recently, transformer-based trackers have become the leading approach, surpassing traditional CNN-based trackers in accuracy. However, their high computational demands hinder their deployment on edge platforms, necessitating efficient solutions. To address this, we propose a novel Neural Architecture Search (NAS) framework designed for transformer-based trackers, named NASTrack. This framework incorporates token pruning to optimize both transformer block structures and the layer-wise token keeping ratio, striking a balance between performance and efficiency. To handle the larger search space introduced by the keeping ratio, we propose a blacklist strategy and a matching-based distillation driven by the Token Overlap Ratio (TOR). Our method discovers hundreds of high-performing trackers, with FLOPs ranging from 1G to 18G. The searched trackers consistently outperform existing efficient state-of-the-art trackers such as CompressTracker and LiteTrack under comparable computational budgets.The code and models are available at https://github.com/Cyhoon84/NASTrack.git.
Tool-augmented methods aim to enhance the reasoning capabilities of large language models (LLMs) by invoking external tools, which can be broadly categorized into training-free and training-based methods. Training-free methods can directly instruct LLMs to invoke external tools, but they exhibit limited generalization in dynamic environments. In contrast, training-based methods use a teacher model to generate tool-use trajectories and fine-tune a student model to distill the trajectories that can generalize across tasks. However, they still face two major challenges: (1) Knowledge boundary mismatch: teacher trajectories may contain steps that the student cannot execute without tools due to their mismatch of knowledge boundaries. (2) Tool overuse propagation: the student inherits the misaligned tool-use strategy from teacher model. To address these challenges, we propose RADAR, an automated framework for knowledge boundary discovery and tool overuse mitigation. First, RADAR performs semantic anchoring to select representative seeds. Second, simulated annealing probing is employed to actively explore the student model’s tool-use boundary and acquire sparse, verified student-aware labels. Third, it globally propagates these labels and rewrites conflicts, yielding deployment-aligned decisions with fewer unnecessary tool calls. We conduct experiments on multiple benchmarks, compared to SOTA method, RADAR reduces average tool-use by 52.7% while improving accuracy by 3.2% across three models.
Systematic safety evaluation of large language models must uncover diverse policy violations under tight query budgets. However, most red-teaming methods optimize attack success rate and repeatedly probe a narrow set of vulnerabilities, yielding redundant failures and leaving rarer yet critical violation categories unexplored. Under fixed budgets, such inefficient exploration delays the first discovery and limits category coverage. To address these limitations, we propose the Budget-Efficient Adaptive Cognitive Offense Navigator (BEACON), a budget-aware safety testing framework that uses Cognitive-Guided Monte Carlo Tree Search to navigate the violation search space under fixed budgets. BEACON innovatively approaches safety testing as a budget-constrained failure discovery process, aiming to identify diverse safety violations as early as possible within a fixed query budget. It also provides an efficiency-oriented evaluation perspective that measures early discovery and harm category coverage under budget constraints. Experiments on standard benchmarks and frontier LLMs show that BEACON discovers failures earlier and achieves higher coverage across policy violation categories. These results underscore the value of evaluating safety testing through discovery efficiency rather than attack success rate alone. Warning: This paper contains examples of harmful language and images, and reader discretion is recommended.
We initiate the study of online contracts, which integrate the game-theoretic considerations of economic contract theory, with the algorithmic and informational challenges of online algorithm design. Our starting point is the classic online setting with preemption, in which a hiring principal faces a sequence of adversarial agent arrivals. Upon arrival, the principal must decide whether to tentatively accept the agent to their team, and whether to dismiss previous tentative choices. Dismissal is irrevocable, giving the setting its online decision-making flavor. In our setting, the agents are rational players: once the team is finalized, a game is played where the principal offers contracts (performance-based payment schemes), and each agent decides whether or not to work. Working agents reward the principal, and the goal is to choose a team that maximizes the principal's utility. Our main positive result is a 1/2-competitive algorithm when agent rewards are additive, which matches the best-possible competitive ratio. Our algorithm is randomized and this is necessary, as we show that no deterministic algorithm can attain a bounded competitive ratio. Moreover, if agent rewards are allowed to exhibit combinatorial structure known as XOS, even randomized algorithms might fail. En route to our competitive algorithm, we develop the technique of balance points, which can be useful for further exploration of online contracts in the adversarial model.
Reconstruction-based frameworks are widely adopted in Time Series Anomaly Detection (TSAD), assuming that models reconstruct normal behavior well but yield larger errors on anomalies. However, in unsupervised TSAD, minimizing reconstruction loss alone often breaks this assumption. Models tend to overfit local patterns for trivial reconstruction and fail to capture the global dependencies that characterize normal behavior. Consequently, anomalies that violate global dependencies can also be reconstructed well, leading to a misalignment between reconstruction and detection. To address this challenge, we propose AnoMamba, a novel TSAD framework that aligns reconstruction with anomaly detection by enhancing global dependency modeling. Specifically, AnoMamba employs patch embedding to reduce local redundancy and introduces a Mamba variant with global step-size reweighting to select meaningful global dependencies. The reweighting process is guided by multi-scale long-tail priors, which adaptively balance global and local dependencies to mitigate overfitting in the unsupervised setting. Extensive experiments on both univariate and multivariate benchmarks demonstrate that AnoMamba consistently outperforms state-of-the-art methods in both accuracy and efficiency, while offering interpretability through its hidden attention map.
Social recommendation leverages social relations to enhance user preference modeling. However, real-world social networks are often noisy and unreliable, where misleading social relationships introduce anisotropic perturbations into user representations. Existing denoising approaches, including heuristic filtering and generative reconstruction, struggle to produce social embeddings that are well aligned with user preferences, limiting their effectiveness in downstream recommendation tasks. To address this challenge, we propose Manifold-Aligned Rectification Flow (MARF), a flow matching based framework that explicitly rectifies noisy social representations toward preference-aligned embeddings. MARF jointly integrates social relations and user-item interactions to construct a preference-aware manifold as the target distribution, guiding the learning of a continuous vector field that captures preference-oriented transformations in the social domain. Through this learned transport process, MARF effectively bridges the gap between the social and preference domains, yielding robust and discriminative user representations. Extensive experiments demonstrate that our proposed model consistently outperforms state-of-the-art social recommendation methods, particularly under sparse and noisy conditions, validating its effectiveness and robustness.
Subgame solving is a technique for scaling algorithms to large games by locally refining a precomputed blueprint strategy during gameplay. While straightforward in perfect-information games where search starts from the current state, subgame solving in imperfect-information games must account for hidden states and uncertainty about the opponent's past strategy. Gadget games were developed to ensure that the improved subgame strategy is robust against any possible opponent's strategy in a zero-sum game. Gadget games typically contain infinitely many Nash equilibria. We demonstrate that while these equilibria are equivalent in the gadget game, they yield vastly different performance in the full game, even when facing a rational opponent. We propose gadget game sequential equilibria as the preferred solution concept. We introduce modifications to the sequence-form linear program and counterfactual regret minimization that converge to these refined solutions with only mild additional computational cost. Additionally, we provide several new insights into the surprising superiority of the resolving gadget game over the max-margin gadget game. Our experiments compare different Nash equilibria of gadget games in several standard benchmark games, showing that our refined equilibria consistently outperform unrefined Nash equilibria, and can reduce the exploitability of the overall strategy by more than 50%.
Physical-layer device identification serves as a critical security mechanism for mitigating cyberattacks and enhancing network resilience. However, its deployment in modern full-duplex Ethernet networks faces two major challenges: (1) the presence of Out-of-Distribution (OOD) data caused by temporal distribution shift and diverse local transmitters, and (2) the difficulty of environment-agnostic device identification, where the unique features of target devices are often obscured in mixed signals. To address these challenges, we propose GenID, which consists of two core components. First, GenID constructs prototypes to emulate the full communication environment through a two stage process: offline and online. Based on the prototype context, it then introduces an invariant feature extraction method to disentangle mixed signals and improve generalization under OOD conditions.Extensive experiments demonstrate that GenID significantly outperforms state-of-the-art baselines, achieving an increase of at least 7.94\% in Macro-F1 score under OOD environments. Our code is available on https://github.com/zyw2004/GenID_IJCAI2026.
Current self-evolution methods mostly chase recall, they learn from any trajectory that ends with the right answer, even when the path relies on a "lucky guess". Such reasoning appears valid in-distribution but often fails the moment the task shifts slightly. We argue that robust evolution implies Structural Invariance: a reasoning path is valid only if its core dependency graph remains isomorphic under counterfactual perturbations. I-EDI enforces this with Verifiable Counterfactual Simulation (VCS). Instead of treating data generation as mere "sample and filter," we treat it as an intervention. Unlike standard rejection sampling, VCS acts as a structural stress test: it accepts a reasoning trace only if it remains valid across generated counterfactuals (e.g., numerical perturbations, context shifts). While this strict filtering reduces the volume of training data (trading recall for precision), our experiments on MATH and GSM8K show it effectively prevents hallucination accumulation, achieving superior out-of-distribution generalization compared to baselines.
The increasing availability of trajectory data is often hampered by sparsity and noise. Existing trajectory recovery methods are further limited by either information loss from coordinate discretization into location IDs, or the inefficiency of conventional diffusion models that require a lengthy denoising process from pure noise. To address these challenges, we propose DiffVec, a novel and efficient diffusion framework for free-space trajectory recovery that operates directly on continuous coordinate data. The backbone of our framework is MVformer, a Transformer-based architecture that models motion vectors—the differences between consecutive coordinates—to better capture the local motion patterns of movement. This model is trained within our ResTraj diffusion paradigm, which commences the denoising process from a structured, interpolated prior to significantly reduce sampling steps. Extensive experiments on the Geolife and Porto datasets demonstrate that DiffVec significantly outperforms state-of-the-art baselines across MAE, MSE, and NDTW.
Deep incomplete multi-view clustering methods can mine patterns of incomplete multi-view data without labels, gaining great attention in various domains. However, current methods are obsessed with aligning view-specific representations from available samples with complete views to learn view-invariant information for missing data recovery, ignoring fruitful view-complementary information beneficial to data imputations. Meanwhile, such the view-invariant learning often leads to the problem of dominant view dependency that models tend to over-rely on views with clear clustering structures and neglect weaker ones, causing a performance bottleneck. Therefore, a content-style disentanglement guided representation learning is proposed for the incomplete multi-view clustering (CSMVC). Specifically, CSMVC designs a view-specific dual-representation learning architecture that extracts view-invariant content representations and view-specific style representations from self-supervised reconstructions of each view with the guidance of the Hilbert schmidt independence criterion. Then, it performs a content-centered style-modulated representation imputation mechanism to infer missing data via fully utilizing inter-view complementary and consistent information. Meanwhile, it devises a cross-view consensus structure mining strategy to capture the clustering structure with intra-cluster compactness and inter-cluster separability from attention-based fusion representations that adaptively aggregate content representations and style representations of each view. Finally, comprehensive experiments show that CSMVC outperforms state-of-the-art methods.
Predicting cross-sectional stock returns is challenging due to low signal-to-noise ratios and evolving market regimes. Classical factor models offer interpretability but limited flexibility, while deep learning models achieve strong performance yet often underutilize financial priors. We address this gap with PRISM-VQ (PRior-Informed Stock Model with Vector Quantization), a dynamic factor framework that integrates expert prior factors, vector-quantized discrete latent factors learned from cross-sectional structure, and a structure-conditioned Mixture-of-Experts to generate time-varying factor loadings. Vector quantization acts as an information bottleneck that suppresses noise while capturing robust market structure, with discrete codes serving both as latent factors and as routing signals for temporal expert specialization. Experiments on CSI 300 and S&P 500 show consistent improvements in cross-sectional return prediction and portfolio performance over strong baselines while preserving interpretability. Our code is available at https://github.com/finxlab/PRISM-VQ.
Fully automatic camera movement directly affects the art quality of dance expressiveness, especially in terms of visual expression, as well as choreography and music. Current studies mainly focus on synthesizing camera movements conditioned on dance and music, but they overlook the camera movement style, which is essential factor for artistic and visual coherence. In this paper, we introduce DanceStyleCam, a unified framework that incorporates the style-consistent characteristic into dance camera movement synthesis with diverse stylistic characteristics. Specifically, a style-aware feature learning module is proposed to map dance style information into compact embeddings, facilitating stable and discriminative style learning. To further guarantee that the generated camera movements remain faithful to the target style, we propose a style-consistent adversarial training scheme, leading and optimizing the model to learn better style-consistent representations. In addition, we also enrich the DCM dataset with diverse camera movement style annotations. Extensive experiments demonstrate that DanceStyleCam outperforms state-of-the-art methods in both generation quality and style consistency. Project page: https://github.com/YCCCCM/DanceStyleCam.
We address the problem of selecting k representative nodes from a network, aiming to simultaneously achieve two objectives: identifying the most influential nodes and ensuring that the selection proportionally reflects the diversity within the network. We propose a general approach to accomplish this by combining ideas from network science and computational social choice. Notably, our algorithms depend only on the connections between nodes and do not utilize any additional information that would explicitly identify groups of nodes. We analyze them theoretically, and demonstrate their effectiveness through a series of experiments.