robustness
The capacity of an AI model to maintain performance in the presence of noise, adversarial attacks, or variations in input data, highlighting its reliability.
- $\Delta \mathrm{Energy}$: Optimizing Energy Change During Vision-Language Alignment Improves both OOD Detection and OOD Generalization
- A Near-optimal, Scalable and Parallelizable Framework for Stochastic Bandits Robust to Adversarial Corruptions and Beyond
- A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
- AGI-Elo: How Far Are We From Mastering A Task?
- AHa-Bench: Benchmarking Audio Hallucinations in Large Audio-Language Models
- AVCD: Mitigating Hallucinations in Audio-Visual Large Language Models through Contrastive Decoding
- Activation-Informed Merging of Large Language Models
- Adaptive Re-calibration Learning for Balanced Multimodal Intention Recognition
- Adaptive Variance Inflation in Thompson Sampling: Efficiency, Safety, Robustness, and Beyond
- Adversarial Diffusion for Robust Reinforcement Learning
- AgentTTS: Large Language Model Agent for Test-time Compute-optimal Scaling Strategy in Complex Tasks
- Aligning Evaluation with Clinical Priorities: Calibration, Label Shift, and Error Costs
- An Analysis of Concept Bottleneck Models: Measuring, Understanding, and Mitigating the Impact of Noisy Annotations
- An Evidence-Based Post-Hoc Adjustment Framework for Anomaly Detection Under Data Contamination
- Assessing the quality of denoising diffusion models in Wasserstein distance: noisy score and optimal bounds
- Attribution-Driven Adaptive Token Pruning for Transformers
- Availability-aware Sensor Fusion via Unified Canonical Space
- Backdoor Mitigation via Invertible Pruning Masks
- Bayesian Optimization with Preference Exploration using a Monotonic Neural Network Ensemble
- Boosting Resilience of Large Language Models through Causality-Driven Robust Optimization
- Boosting the Uniqueness of Neural Networks Fingerprints with Informative Triggers
- Bounds on the computational complexity of neurons due to dendritic morphology
- Breaking the Gradient Barrier: Unveiling Large Language Models for Strategic Classification
- Broken Tokens? Your Language Model can Secretly Handle Non-Canonical Tokenizations
- CLEAR: Command Level Annotated Dataset for Ransomware Detection
- Capturing Polysemanticity with PRISM: A Multi-Concept Feature Description Framework
- Certifying Deep Network Risks and Individual Predictions with PAC-Bayes Loss via Localized Priors
- ChA-MAEViT: Unifying Channel-Aware Masked Autoencoders and Multi-Channel Vision Transformers for Improved Cross-Channel Learning
- Co-Regularization Enhances Knowledge Transfer in High Dimensions
- ColorBench: Can VLMs See and Understand the Colorful World? A Comprehensive Benchmark for Color Perception, Reasoning, and Robustness
- Continuous Domain Generalization
- Curvature Tuning: Provable Training-free Model Steering From a Single Parameter
- DCI: Dual-Conditional Inversion for Boosting Diffusion-Based Image Editing
- DETree: DEtecting Human-AI Collaborative Texts via Tree-Structured Hierarchical Representation Learning
- DGCBench: A Deep Graph Clustering Benchmark
- DINO-Foresight: Looking into the Future with DINO
- DIPO: Dual-State Images Controlled Articulated Object Generation Powered by Diverse Data
- DPAIL: Training Diffusion Policy for Adversarial Imitation Learning without Policy Optimization
- DeblurDiff: Real-Word Image Deblurring with Generative Diffusion Models
- Demystifying Network Foundation Models
- Differentiation Through Black-Box Quadratic Programming Solvers
- Diffusion Guided Adversarial State Perturbations in Reinforcement Learning
- Directed-Tokens: A Robust Multi-Modality Alignment Approach to Large Language-Vision Models
- Discovering Symbolic Partial Differential Equation by Abductive Learning
- Distillation Robustifies Unlearning
- Distributional Adversarial Attacks and Training in Deep Hedging
- Doodle to Detect: A Goofy but Powerful Approach to Skeleton-based Hand Gesture Recognition
- DualFocus: Depth from Focus with Spatio-Focal Dual Variational Constraints
- DynaGuide: Steering Diffusion Polices with Active Dynamic Guidance
- Dynamic Semantic-Aware Correlation Modeling for UAV Tracking
- Efficient Multi-modal Large Language Models via Progressive Consistency Distillation
- Elastic Robust Unlearning of Specific Knowledge in Large Language Models
- Embracing Contradiction: Theoretical Inconsistency Will Not Impede the Road of Building Responsible AI Systems
- Empirical Study on Robustness and Resilience in Cooperative Multi-Agent Reinforcement Learning
- Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attacks
- Enhancing Optimizer Stability: Momentum Adaptation of The NGN Step-size
- ErrorTrace: A Black-Box Traceability Mechanism Based on Model Family Error Space
- Evaluating and Learning Optimal Dynamic Treatment Regimes under Truncation by Death
- Factor Decorrelation Enhanced Data Removal from Deep Predictive Models
- Feature-aware Modulation for Learning from Temporal Tabular Data
- FedGPS: Statistical Rectification Against Data Heterogeneity in Federated Learning
- FedRW: Efficient Privacy-Preserving Data Reweighting for Enhancing Federated Learning of Language Models
- ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation Detection
- Fortifying Time Series: DTW-Certified Robust Anomaly Detection
- From Cradle to Cane: A Two-Pass Framework for High-Fidelity Lifespan Face Aging
- From Experts to a Generalist: Toward General Whole-Body Control for Humanoid Robots
- From Faults to Features: Pretraining to Learn Robust Representations against Sensor Failures
- From Pretraining to Pathology: How Noise Leads to Catastrophic Inheritance in Medical Models
- GMV: A Unified and Efficient Graph Multi-View Learning Framework
- GaRA-SAM: Robustifying Segment Anything Model with Gated-Rank Adaptation
- Gymnasium: A Standard Interface for Reinforcement Learning Environments
- H-SPLID: HSIC-based Saliency Preserving Latent Information Decomposition
- HAODiff: Human-Aware One-Step Diffusion via Dual-Prompt Guidance
- HYPERION: Fine-Grained Hypersphere Alignment for Robust Federated Graph Learning
- HiMoLE: Towards OOD-Robust LoRA via Hierarchical Mixture of Experts
- Human Texts Are Outliers: Detecting LLM-generated Texts via Out-of-distribution Detection
- Human-assisted Robotic Policy Refinement via Action Preference Optimization
- Hyper-Modality Enhancement for Multimodal Sentiment Analysis with Missing Modalities
- IPAD: Inverse Prompt for AI Detection - A Robust and Interpretable LLM-Generated Text Detector
- Image Super-Resolution with Guarantees via Conformalized Generative Models
- Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing
- InFlux: A Benchmark for Self-Calibration of Dynamic Intrinsics of Video Cameras
- Influence Functions for Edge Edits in Non-Convex Graph Neural Networks
- Is Artificial Intelligence Generated Image Detection a Solved Problem?
- Is Limited Participant Diversity Impeding EEG-based Machine Learning?
- Keeping an Eye on LLM Unlearning: The Hidden Risk and Remedy
- LLM Unlearning via Neural Activation Redirection
- LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders
- Language‑Bias‑Resilient Visual Question Answering via Adaptive Multi‑Margin Collaborative Debiasing
- Latent Space Factorization in LoRA
- Learning (Approximately) Equivariant Networks via Constrained Optimization
- Learning (Approximately) Equivariant Networks via Constrained Optimization
- Learning Provably Improves the Convergence of Gradient Descent
- Learning-Augmented Online Bidding in Stochastic Settings
- Less Greedy Equivalence Search
- Lifelong Safety Alignment for Language Models
- MARS-VFL: A Unified Benchmark for Vertical Federated Learning with Realistic Evaluation
- MIR-Bench: Can Your LLM Recognize Complicated Patterns via Many-Shot In-Context Reasoning?
- MIX: A Multi-view Time-Frequency Interactive Explanation Framework for Time Series Classification
- ML4CFD Competition: Results and Retrospective Analysis
- MOTION: Multi-Sculpt Evolutionary Coarsening for Federated Continual Graph Learning
- MS-Bench: Evaluating LMMs in Ancient Manuscript Study through a Dunhuang Case Study
- Mask Image Watermarking
- Measuring what Matters: Construct Validity in Large Language Model Benchmarks
- Mechanism Design for LLM Fine-tuning with Multiple Reward Models
- Metric Automata Theory: A Unifying Theory of RNNs
- Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering
- Mitigating Spurious Features in Contrastive Learning with Spectral Regularization
- MixAT: Combining Continuous and Discrete Adversarial Training for LLMs
- Modality-Aware SAM: Sharpness-Aware-Minimization Driven Gradient Modulation for Harmonized Multimodal Learning
- Multi-Agent Reinforcement Learning with Communication-Constrained Priors
- Multivariate Time Series Anomaly Detection with Idempotent Reconstruction
- NoisyGRPO: Incentivizing Multimodal CoT Reasoning via Noise Injection and Bayesian Estimation
- Normalize Filters! Classical Wisdom for Deep Vision
- On the Empirical Power of Goodness-of-Fit Tests in Watermark Detection
- On the Robustness of Verbal Confidence of LLMs in Adversarial Attacks
- On the Stability of Graph Convolutional Neural Networks: A Probabilistic Perspective
- Optimal Control for Transformer Architectures: Enhancing Generalization, Robustness and Efficiency
- Optimizing Anytime Reasoning via Budget Relative Policy Optimization
- Orthogonal Survival Learners for Estimating Heterogeneous Treatment Effects from Time-to-Event Data
- Pattern-Guided Adaptive Prior for Structure Learning
- Position: AI Should Sense Better, Not Just Scale Bigger: Adaptive Sensing as a Paradigm Shift
- Probabilistic Stability Guarantees for Feature Attributions
- Problem-Parameter-Free Decentralized Bilevel Optimization
- Program Synthesis via Test-Time Transduction
- Provable Ordering and Continuity in Vision-Language Pretraining for Generalizable Embodied Agents
- REOBench: Benchmarking Robustness of Earth Observation Foundation Models
- RL Tango: Reinforcing Generator and Verifier Together for Language Reasoning
- ReMA: Learning to Meta-Think for LLMs with Multi-agent Reinforcement Learning
- Reaction Prediction via Interaction Modeling of Symmetric Difference Shingle Sets
- Reasoning Is Not a Race: When Stopping Early Beats Going Deeper
- Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval
- Remarkable Robustness of LLMs: Stages of Inference?
- Robust Federated Finetuning of LLMs via Alternating Optimization of LoRA
- Robust Graph Condensation via Classification Complexity Mitigation
- Robust Policy Expansion for Offline-to-Online RL under Diverse Data Corruption
- Robust and Diverse Multi-Agent Learning via Rational Policy Gradient
- SPARTAN: A Sparse Transformer World Model Attending to What Matters
- Safe-Sora: Safe Text-to-Video Generation via Graphical Watermarking
- Salient Concept-Aware Generative Data Augmentation
- Sample-efficient Learning of Concepts with Theoretical Guarantees: from Data to Concepts without Interventions
- Scaling Image Geo-Localization to Continent Level
- Self-Guided Hierarchical Exploration for Generalist Foundation Model Web Agents
- Semi-Supervised Regression with Heteroscedastic Pseudo-Labels
- Set-LLM: A Permutation-Invariant LLM
- Short-length Adversarial Training Helps LLMs Defend Long-length Jailbreak Attacks: Theoretical and Empirical Evidence
- SimSort: A Data-Driven Framework for Spike Sorting by Large-Scale Electrophysiology Simulation
- Simple and Effective Specialized Representations for Fair Classifiers
- Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
- Stackelberg Self-Annotation: A Robust Approach to Data-Efficient LLM Alignment
- State Entropy Regularization for Robust Reinforcement Learning
- State Entropy Regularization for Robust Reinforcement Learning
- Streaming Audio Generation from Discrete Tokens via Streaming Flow Matching
- Structured Spectral Reasoning for Frequency-Adaptive Multimodal Recommendation
- StyleGuard: Preventing Text-to-Image-Model-based Style Mimicry Attacks by Style Perturbations
- THUNDER: Tile-level Histopathology image UNDERstanding benchmark
- TRUST: Test-Time Refinement using Uncertainty-Guided SSM Traverses
- The Best Instruction-Tuning Data are Those That Fit
- The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models
- The Indra Representation Hypothesis
- The Rashomon Set Has It All: Analyzing Trustworthiness of Trees under Multiplicity
- Towards Generalizable 3D Human Pose Estimation via Ensembles on Flat Loss Landscapes
- Towards Prospective Medical Image Reconstruction via Knowledge-Informed Dynamic Optimal Transport
- Towards the Resistance of Neural Network Fingerprinting to Fine-tuning
- Uncertainty Quantification for Physics-Informed Neural Networks with Extended Fiducial Inference
- Vanish into Thin Air: Cross-prompt Universal Adversarial Attacks for SAM2
- VisualLens: Personalization through Task-Agnostic Visual History
- What Really is a Member? Discrediting Membership Inference via Poisoning
- What We Miss Matters: Learning from the Overlooked in Point Cloud Transformers
- Whose Instructions Count? Resolving Preference Bias in Instruction Fine-Tuning