Singapore

Membership inference attacks (MIAs) test whether a data point was part of a model&#39;s training set, posing serious privacy risks. Existing methods often depend on shadow models or heavy query access, which limits their practicality. We propose GP-MIA, an efficient and interpretable approach based on Gaussian process (GP) meta-modeling. Using post-hoc metrics such as accuracy, entropy, dataset statistics, and optional sensitivity features (e.g. gradients, NTK measures) from a single trained model, GP-MIA trains a GP classifier to distinguish members from non-members while providing calibrated uncertainty estimates. Experiments on synthetic data, real-world fraud detection data, CIFAR-10, and WikiText-2 show that GP-MIA achieves high accuracy and generalizability, offering a practical alternative to existing MIAs.

AAAI 2026

Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach

post hoc analysis

membership inference attacks

gaussian process

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

The purpose of the AAAI conference series is to promote research in Artificial Intelligence (AI) and foster scientific exchange between researchers, practitioners, scientists, students, and engineers across the entirety of AI and its affiliated disciplines. AAAI-26 will feature technical paper presentations, special tracks, invited speakers, workshops, tutorials, poster sessions, senior member presentations, competitions, and exhibit programs, and a range of other activities to be announced.<br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to AAAI 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://aaai.org/conference/aaai/aaai-26/registration/) first.

**Online Registration Form**: https://aaai.getregistered.net/conference-2026 

Registration Required

We are pleased to announce the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26), which will be held in Singapore EXPO from January 20 to January 27, 2026.

Membership inference attacks (MIAs) test whether a data point was part of a model's training set, posing serious privacy risks. Existing methods often depend on shadow models or heavy query access, which limits their practicality. We propose GP-MIA, an efficient and interpretable approach based on Gaussian process (GP) meta-modeling. Using post-hoc metrics such as accuracy, entropy, dataset statistics, and optional sensitivity features (e.g. gradients, NTK measures) from a single trained model, GP-MIA trains a GP classifier to distinguish members from non-members while providing calibrated uncertainty estimates. Experiments on synthetic data, real-world fraud detection data, CIFAR-10, and WikiText-2 show that GP-MIA achieves high accuracy and generalizability, offering a practical alternative to existing MIAs.

workshop paper

Recursive Self-Improvement (RSI) is an effective approach to alleviating data scarcity. This paper explores the application of RSI in text-to-image diffusion models, addressing the challenge of training collapse caused by biased synthetic data. We identify two key factors contributing to this collapse: the lack of perceptual alignment and the accumulation of generative hallucinations. To mitigate these issues, we propose three strategies: (1) a prompt construction and filtering pipeline designed to facilitate the generation of perceptual aligned data, (2) a preference sampling method to identify human-preferred samples and filter out generative hallucinations, and (3) a distribution-based weighting scheme to penalize selected samples with hallucinatory errors. Our extensive experiments validate the effectiveness of these approaches.

Generating on Generated: An Approach Towards Self-Evolving Diffusion Models

Machine learning holds great promise for advancing the field of medicine, with electronic health records (EHRs) serving as a primary data source. However, EHRs are often sparse and contain missing data due to various challenges and limitations in data collection and sharing between healthcare providers. Existing techniques for imputing missing values predominantly focus on known unknowns, such as missing or unavailable values of lab test results; most do not explicitly address situations where it is difficult to distinguish what is missing. For instance, a missing diagnosis code in an EHR could signify either that the patient has not been diagnosed with the condition or that a diagnosis was made, but not shared by a provider. Such situations fall into the paradigm of unknown unknowns. To address this challenge, we develop a general purpose algorithm for denoising data to recover unknown missing values in binary EHRs. We design a transformer-based denoising neural network where the output is thresholded adaptively to recover values in cases where we predict data are missing. Our method demonstrates improved accuracy in denoising medical codes within a real EHR dataset compared to existing imputation approaches and leads to increased performance on downstream tasks using the denoised data. In particular, when applying our method to a real world application, predicting hospital readmission from EHRs, our method achieves statistically significant improvement over all existing baselines.

Imputation of Unknown Missingness in Sparse Electronic Health Records

Sequential recommendation (SR) systems leverage users’ historical interaction sequences and are widely adopted on online platforms to deliver personalized suggestions effectively. Recent advancements in self-attention and contrastive learning-based models have significantly improved SR performance. However, these methods often utilize user behavior on a fixed, single time scale, making them insensitive to the temporal evolution of user interests. While these models effectively capture long-term preferences, they are biased toward persistent behavioral patterns and fail to reflect transient, short-term dynamics, as indicated by their minimal performance degradation on shuffled interaction sequences. To address these challenges, we propose Hierarchical INTerests-aware Sequential Recommendation (HINT). Instead of relying on contrastive learning and self-attention methods, HINT implements a hierarchical user interests modeling approach. Specifically, it first forms multiple sequences by applying variable-kernel average pooling to the original sequence, a technique designed to capture user interests across multiple time scales. It then employs a Fourier transform-based filtering to adaptively denoise each view by processing its low- and high-frequency components. Subsequently, HINT uses an attention-based adaptive aggregation mechanism to integrate the multi-scale views. This allows the model to focus on salient information within the hierarchical interests representations, resulting in a more accurate and comprehensive user profile. Extensive experiments conducted on four real-world datasets demonstrate that HINT outperforms other state-of-the-art baselines.

Hierarchical Interests Modeling on Multiple Time Scales for Sequential Recommendation

In data-scarce settings, learning accurate and interpretable
models for tabular classification remains a fundamental chal-
lenge. We propose a novel pipeline for few-shot tabular clas-
sification that distills the predictive power of large founda-
tion models into globally interpretable student models. Given
a tabular dataset, we generate synthetic data using CTGAN
to approximate the underlying distribution. We then train
high-capacity teacher models (TabLLM, TabPFN, TabM) on
a small number of labeled examples and use them to pseu-
dolabel the synthetic data. Finally, we train student explainer
models (i.e., XGBoost, logistic regression, decision trees,
Generalized Linear Rule Model (GLRM), and TTnet) on
this pseudolabeled synthetic dataset. These student models
are globally and exactly interpretable, yielding logical de-
cision rules (e.g., disjunctive normal form) that fully repro-
duce their predictions. Across 15 tabular binary classification
benchmarks, our distilled models generally outperform base-
lines trained directly on the few-shot data, with improved
ROC AUC scores in the low-shot (4-16 samples) to mid-
and higher-shot settings (up to 256 samples). In particular,
TabPFN-based distillation achieves strong and reliable im-
provements across the entire shot spectrum, not only in heavy
data scarcity but also in broader low-data regimes. This work
demonstrates that foundation models can be effectively lever-
aged as teachers to produce small, transparent, and high-
performing classifiers. Our approach advances the goal of re-
liable and interpretable machine learning in real-world set-
tings where labeled data is limited.

Towards Globally Interpretable Few-Shot Tabular Learning: Distilling Foundation Models

'LLM-as-a-judge,'' which utilizes large language models (LLMs) as evaluators, has proven effective in many evaluation tasks.
However, evaluator LLMs exhibit numerical bias, a phenomenon where certain evaluation scores are generated disproportionately often, leading reduced evaluation performance.
This study investigates the cause of this bias.
Given that most evaluator LLMs are aligned through instruction tuning and preference tuning, and that prior research suggests alignment reduces output diversity, we hypothesize that numerical bias arises from alignment.
To test this, we compare outputs from pre- and post-alignment LLMs, and observe that alignment indeed increases numerical bias.
We also explore mitigation strategies for post-alignment LLMs, including temperature scaling, distribution calibration, and score range adjustment.
Among these, score range adjustment is most effective in reducing bias and improving performance, though still heuristic.
Our findings highlight the need for further work on optimal score range selection and more robust mitigation strategies.

Exploring the Effects of Alignment on Numerical Bias in Large Language Models

Adaptations are one of the most prominent methods that enable the utilization of large backbone models for those who have limited computational resources. Yet, even in their last resort of using adaptations, many face a dead end in training when adaptation data are insufficient. Since the acquisition of adaptation data is arduous, the situation frequently appears, daunting without a particular solution. Often, even with extra training or random data augmentation, the performance of adaptations does not improve. To address this, we organize a solid probabilistic background and sampling method that focuses on the enormous amount of backbone data used for pre-training. We name the method, Backbone Augmented Training (BAT). First, we formulate and prove two key probabilistic propositions: one establishes the validity of BAT, while the other identifies a condition under which BAT benefits adaptation. Furthermore, we introduce an original data sampling scheme that satisfies these propositions and present ALgorithm for BAT (ALBAT) to efficiently implement this approach. ALBAT effectively enhances adaptation training even when other methods meet performance stalemate due to adaptation data scarcity.

BAT: Backbone Augmented Training for Adaptations

This paper investigates how the degree of group fairness changes when the degree of individual fairness is actively controlled. As a metric quantifying individual fairness, we consider generalized entropy (GE) recently introduced into machine learning community. To control the degree of individual fairness, we design a classification algorithm satisfying a given degree of individual fairness through an empirical risk minimization (ERM) with a fairness constraint specified in terms of GE. We show the PAC learnability of the fair ERM problem by proving that the true fairness degree does not deviate much from an empirical one with high probability for finite VC dimension if the sample size is big enough. Our experiments show that strengthening individual fairness degree does not always lead to enhancement of group fairness.

Impacts of Individual Fairness on Group Fairness from the Perspective of Generalized Entropy

Semi-supervised object detection (SSOD) has achieved notable progress through advances in self-training frameworks. However, most existing approaches still rely solely on classification confidence to select pseudo-labels, overlooking the localization uncertainty. This results in pseudo-labels that lack sufficient localization information, since classification confidence alone does not adequately represent the accuracy of spatial localization. In this paper, we address this limitation from two complementary perspectives: edge-level reliability estimation and instance-level pseudo-label selection. From the former perspective, we quantify localization confidence using a simple statistic — the maximum adjacent probability sum derived from the Distribution Focal Loss — and aggregate edge-level scores into a box-level reliability via the minimum over edges. This suppresses boxes with at least one uncertain boundary, improving spatial precision with negligible computational overhead. From the latter perspective, we integrate this localization-confidence-aware criterion into Efficient Teacher, a YOLO-oriented semi-supervised object detection framework, to adaptively filter pseudo-labels during self-training. Experiments on the KITTI dataset using YOLOv8-m demonstrate consistent improvements over baselines and improved training stability. Further analysis reveals that our approach achieves higher recall of pseudo-labels while maintaining precision, validating the effectiveness of localization confidence modeling in SSOD.

Localization-Confidence-Aware Pseudo-Label Selection for YOLO-based Semi-Supervised Object Detection

Eye movements during reading provide valuable data for studying cognitive processes and text properties. While eye-tracking data can enhance language models, its application is limited by data scarcity, particularly for low-resource languages. Although synthetic scanpath generation offers a promising solution, current models are largely confined to English and Chinese. This work explores multilingual model for scanpath generation, demonstrating that a single model trained on 13 typologically diverse languages outperforms monolingual counterparts across most metrics. By leveraging shared linguistic patterns, our approach mitigates data scarcity for individual languages and facilitates synthetic gaze data creation for broader multilingual NLP applications.

Cross-Lingual Transfer Learning for Enhanced Synthetic Scanpath Prediction in Reading

Scene Text Editing (STE) is the task of modifying text content in an image while preserving its visual style, such as font, color, and background. While recent diffusion-based approaches have shown improvements in visual quality, key limitations remain: lack of support for low-resource languages, domain gap between synthetic and real data, and the absence of appropriate metrics for evaluating text style preservation. To address these challenges, we propose STELLAR (Scene Text Editor for Low-resource LAnguages and Real-world data). STELLAR enables reliable multilingual editing through a language‑adaptive glyph encoder and a multi‑stage training strategy that first pre‑trains on synthetic data and then fine‑tunes on real images. We also construct a new dataset, STIPLAR(Scene Text Image Pairs of Low-resource lAnguages and Real-world data), for training and evaluation. Furthermore, we propose Text Appearance Similarity (TAS), a novel metric that assesses style preservation by independently measuring font, color, and background similarity, enabling robust evaluation even without ground truth. Experimental results demonstrate that STELLAR outperforms state-of-the-art models in visual consistency and recognition accuracy, achieving an average TAS improvement of 2.2% across languages over the baselines.

Downloads

Next from AAAI 2026

Generating on Generated: An Approach Towards Self-Evolving Diffusion Models

Similar lecture

Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2026

Generating on Generated: An Approach Towards Self-Evolving Diffusion Models

Similar lecture

Training data membership inference via Gaussian process meta-modeling: a post-hoc analysis approach

Downloads