AAAI-2024-Papers Application Safe, Robust and Responsible AI 🆔 Title Repo Paper Video ImageCaptioner2: Image Captioner for Image Captioning Bias Amplification Assessment ➖ A Framework for Data-Driven Explainability in Mathematical Optimization ➖ On the Importance of Application-Grounded Experimental Design for Evaluating Explainable ML Methods ➖ Risk-Aware Continuous Control with Neural Contextual Bandits ➖ Robust Uncertainty Quantification Using Conformalised Monte Carlo Prediction ➖ CCTR: Calibrating Trajectory Prediction for Uncertainty-Aware Motion Planning in Autonomous Driving ➖ Rethinking the Development of Large Language Models from the Causal Perspective: A Legal Text Prediction Case Study ➖ Truth Forest: Toward Multi-Scale Truthfulness in Large Language Models through Intervention without Tuning ➖ Constrained Meta-Reinforcement Learning for Adaptable Safety Guarantee with Differentiable Convex Programming ➖ Conformal Prediction Regions for Time Series Using Linear Complementarity Programming ➖ TTTS: Tree Test Time Simulation for Enhancing Decision Tree Robustness against Adversarial Examples ➖ Find the Lady: Permutation and Re-synchronization of Deep Neural Networks ➖ Stability Analysis of Switched Linear Systems with Neural Lyapunov Functions ➖ Robustness Verification of Multi-Class Tree Ensembles ➖ P2BPO: Permeable Penalty Barrier-Based Policy Optimization for Safe RL ➖ Trade-Offs in Fine-Tuned Diffusion Models between Accuracy and Interpretability ➖ From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space ➖ Automatically Testing Functional Properties of Code Translation Models ➖ A Simple and Yet Fairly Effective Defense for Graph Neural Networks ➖ Invisible Backdoor Attack against 3D Point Cloud Classifier in Graph Spectral Domain ➖ CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection ➖ Solving Non-rectangular Reward-Robust MDPs via Frequency Regularization ➖ Balance Reward and Safety Optimization for Safe Reinforcement Learning: A Perspective of Gradient Manipulation ➖ π-Light: Programmatic Interpretable Reinforcement Learning for Resource-Limited Traffic Signal Control ➖ Generative Model for Decision Trees ➖ Omega-Regular Decision Processes ➖ Provable Robustness against a Union of L_0 Adversarial Attacks ➖ All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Models ➖ Towards Efficient Verification of Quantized Neural Networks ➖ On the Concept Trustworthiness in Concept Bottleneck Models ➖ Personalization as a Shortcut for Few-Shot Backdoor Attack against Text-to-Image Diffusion Models ➖ Stronger and Transferable Node Injection Attacks ➖ Learning Fair Policies for Multi-Stage Selection Problems from Observational Data ➖ NeRFail: Neural Radiance Fields-Based Multiview Adversarial Attack ➖ Analysis of Differentially Private Synthetic Data: A Measurement Error Approach ➖ Chasing Fairness in Graphs: A GNN Architecture Perspective ➖ Assume-Guarantee Reinforcement Learning ➖ DeepBern-Nets: Taming the Complexity of Certifying Neural Networks Using Bernstein Polynomial Activations and Precise Bound Propagation ➖ Layer Attack Unlearning: Fast and Accurate Machine Unlearning via Layer Level Attack and Knowledge Distillation ➖ Quilt: Robust Data Segment Selection against Concept Drifts ➖ OUTFOX: LLM-Generated Essay Detection Through In-Context Learning with Adversarially Generated Examples ➖ Accelerating Adversarially Robust Model Selection for Deep Neural Networks via Racing ➖ Robust Active Measuring under Model Uncertainty ➖ Towards Large Certified Radius in Randomized Smoothing Using Quasiconcave Optimization ➖ Contrastive Credibility Propagation for Reliable Semi-supervised Learning ➖ Exponent Relaxation of Polynomial Zonotopes and Its Applications in Formal Neural Network Verification ➖ I Prefer Not to Say: Protecting User Consent in Models with Optional Personal Data ➖ Promoting Counterfactual Robustness through Diversity ➖ Revisiting the Information Capacity of Neural Network Watermarks: Upper Bound Estimation and Beyond ➖ PointCVaR: Risk-Optimized Outlier Removal for Robust 3D Point Cloud Classification ➖ Game-Theoretic Unlearnable Example Generator ➖ Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learning ➖ Handling Long and Richly Constrained Tasks through Constrained Hierarchical Reinforcement Learning ➖ Combining Graph Transformers Based Multi-Label Active Learning and Informative Data Augmentation for Chest Xray Classification ➖ Enumerating Safe Regions in Deep Neural Networks with Provable Probabilistic Guarantees ➖ Divide-and-Aggregate Learning for Evaluating Performance on Unlabeled Data ➖ SentinelLMs: Encrypted Input Adaptation and Fine-Tuning of Language Models for Private and Secure Inference ➖ Safeguarded Progress in Reinforcement Learning: Safe Bayesian Exploration for Control Policy Synthesis ➖ Feature Unlearning for Pre-trained GANs and VAEs ➖ Reward Certification for Policy Smoothed Reinforcement Learning ➖ EncryIP: A Practical Encryption-Based Framework for Model Intellectual Property Protection ➖ Neural Closure Certificates ➖ SocialStigmaQA: A Benchmark to Uncover Stigma Amplification in Generative Language Models ➖ MaxEnt Loss: Constrained Maximum Entropy for Calibration under Out-of-Distribution Shift ➖ ORES: Open-Vocabulary Responsible Visual Synthesis ➖ Q-SENN: Quantized Self-Explaining Neural Networks ➖ Understanding Likelihood of Normalizing Flow and Image Complexity through the Lens of Out-of-Distribution Detection ➖ Adversarial Initialization with Universal Adversarial Perturbation: A New Approach to Fast Adversarial Training ➖ A PAC Learning Algorithm for LTL and Omega-Regular Objectives in MDPs ➖ Robust Stochastic Graph Generator for Counterfactual Explanations ➖ Visual Adversarial Examples Jailbreak Aligned Large Language Models ➖ Dissenting Explanations: Leveraging Disagreement to Reduce Model Overreliance ➖ I-CEE: Tailoring Explanations of Image Classification Models to User Expertise ➖ A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy ➖ Interpretability Benchmark for Evaluating Spatial Misalignment of Prototypical Parts Explanations ➖ Human-Guided Moral Decision Making in Text-Based Games ➖ Towards Fairer Centroids in K-means Clustering ➖ Toward Robustness in Multi-Label Classification: A Data Augmentation Strategy against Imbalance and Noise ➖ Bidirectional Contrastive Split Learning for Visual Question Answering ➖ Quantile-Based Maximum Likelihood Training for Outlier Detection ➖ Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention ➖ Toward More Generalized Malicious URL Detection Models ➖ Self-Supervised Likelihood Estimation with Energy Guidance for Anomaly Segmentation in Urban Scenes ➖ Pure-Past Action Masking ➖ Long-Term Safe Reinforcement Learning with Binary Feedback ➖ Identifying Reasons for Bias: An Argumentation-Based Approach ➖ Would You Like Your Data to Be Trained? A User Controllable Recommendation Framework ➖ Moderate Message Passing Improves Calibration: A Universal Way to Mitigate Confidence Bias in Graph Neural Networks ➖ Generating Diagnostic and Actionable Explanations for Fair Graph Neural Networks ➖ Physics-Informed Representation and Learning: Control and Risk Quantification ➖ Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration ➖ Concealing Sensitive Samples against Gradient Leakage in Federated Learning ➖ The Evidence Contraction Issue in Deep Evidential Regression: Discussion and Solution ➖ Byzantine-Robust Decentralized Learning via Remove-then-Clip Aggregation ➖ Hypothesis Testing for Class-Conditional Noise Using Local Maximum Likelihood ➖ Providing Fair Recourse over Plausible Groups ➖ Representation-Based Robustness in Goal-Conditioned Reinforcement Learning ➖ Enhancing Off-Policy Constrained Reinforcement Learning through Adaptive Ensemble C Estimation ➖ Efficient Toxic Content Detection by Bootstrapping and Distilling Large Language Models ➖ LR-XFL: Logical Reasoning-Based Explainable Federated Learning ➖ GaLileo: General Linear Relaxation Framework for Tightening Robustness Certification of Transformers ➖ A Huber Loss Minimization Approach to Byzantine Robust Federated Learning ➖ Responsible Bandit Learning via Privacy-Protected Mean-Volatility Utility ➖ UMA: Facilitating Backdoor Scanning via Unlearning-Based Model Ablation ➖ AdvST: Revisiting Data Augmentations for Single Domain Generalization ➖ Can LLM Replace Stack Overflow? A Study on Robustness and Reliability of Large Language Model Code Generation ➖ DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models ➖ Closing the Gap: Achieving Better Accuracy-Robustness Tradeoffs against Query-Based Attacks ➖ Coevolutionary Algorithm for Building Robust Decision Trees under Minimax Regret ➖
