R
Published on

CVPR 2026 — Adversarial Robustness & Security

Adversarial Robustness & Security

248 papers

1. HeroGS: Hierarchical Guidance for Robust 3D Gaussian Splatting under Sparse Views

2. SAM 3D Body: Robust Full-Body Human Mesh Recovery

3. CLP: A Real-World Dataset of Contaminated Lens Protectors for Robust Semantic Segmentation

4. Computation and Communication Efficient Federated Unlearning via On-server Gradient Conflict Mitigation and Expression

5. The Devil Is in Gradient Entanglement: Energy-Aware Gradient Coordinator for Robust Generalized Category Discovery

6. Mitigating Error Amplification in Fast Adversarial Training

7. RaPA: Enhancing Transferable Targeted Attacks via Random Parameter Pruning

8. Black-box Membership Inference Attacks on the Pre-training Data of Image-generation Models

9. Adversarial Style Optimization: Enhancing VLM Jailbreaks by GRPO-based Stylistic Triggers Optimization

10. When LoRA Betrays: Backdooring Text-to-Image Models by Masquerading as Benign Adapters

11. GDFA: Geometry-Driven Federated Unlearning with Directional Task Vector Alignment

12. All in One: Unifying Deepfake Detection, Tampering Localization, and Source Tracing with a Robust Landmark-Identity Watermark

13. Robust Remote Sensing Image-Text Retrieval with Noisy Correspondence

14. The Image as Its Own Reward: Reinforcement Learning with Adversarial Reward for Image Generation

15. FedSDR: Federated Graph Learning with Structural Noise Detection and Reconstruction

16. Percept-WAM: Perception-Enhanced World-Awareness-Action Model for Robust End-to-End Autonomous Driving

17. Eliminate Distance Differences Induced by Backdoor Attacks: Layer-Selective Training and Clipping to Mask Backdoor Models

18. Attack for Defense: Adversarial Agents for Point Prompt Optimization Empowering Segment Anything Model

19. X-AVDT: Audio-Visual Cross-Attention for Robust Deepfake Detection

20. Rel-Zero: Harnessing Patch-Pair Invariance for Robust Zero-Watermarking Against AI Editing

21. SPDMark: Selective Parameter Displacement for Robust Video Watermarking

22. Air-Know: Arbiter-Calibrated Knowledge-Internalizing Robust Network for Composed Image Retrieval

23. A Combination of Noise and Bilateral Filters Achieve Supralinear and Scalable Adversarial Robustness in CNNs

24. FedARA: Resource-adaptive Low-rank Personalized Federated Learning via Anchor-driven Representation Alignment on Heterogeneous Edge Devices

25. Towards Cross-Modal Preservation, Consistency and Alignment for Privacy-Preserving Visible-Infrared Person Re-Identification

26. Out of Sight, Out of Track: Adversarial Attacks on Propagation-based Multi-Object Trackers via Query State Manipulation

27. Enhancing Visual Representation with Textual Semantics: Textual Semantics-Powered Prototypes for Heterogeneous Federated Learning

28. Proof-of-Perception: Certified Tool-Using Multimodal Reasoning with Compositional Conformal Guarantees

29. RHCNet: Residual-Guided Hierarchical Calibration Network for Robust Underwater Object Detection

30. Diffusion-Based Native Adversarial Synthesis for Enhanced Medical Segmentation Generalization

31. Do Vision-Language Models Leak What They Learn? Adaptive Token-Weighted Model Inversion Attacks

32. Revisiting Geometric Obfuscation with Dual Convergent Lines for Privacy-Preserving Image Queries in Visual Localization

33. All Vehicles Can Lie: Efficient Adversarial Defense in Fully Untrusted-Vehicle Collaborative Perception via Pseudo-Random Bayesian Inference

34. C-LaV: Conditional Latent Velocity Field Denoising for Weather-Robust LiDAR Place Recognition

35. PGA: Prior-free Generative Attack for Practical No-box Scenario

36. FedDAP: Domain-Aware Prototype Learning for Federated Learning under Domain Shift

37. MambaLiteUNet: Cross-Gated Adaptive Feature Fusion for Robust Skin Lesion Segmentation

38. Physical Adversarial Clothing Evades Visible-Thermal Detectors via Non-Overlapping RGB-T Pattern

39. FedHarmony: Harmonizing Heterogeneous Label Correlations in Federated Multi-Label Learning

40. RevINN: An End-to-End Invertible Neural Network for Reversible Adversarial Examples Generation

41. Towards Human-Imperceptible Backdoor Attacks on Text-to-Image Diffusion Models

42. FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction

43. LaSM: Layer-wise Scaling Mechanism for Defending Pop-up Attack on GUI Agents

44. AGiLe: Learning Robust Long-Horizon Manipulation via Affordance-Grounded Bidirectional Latent Planning

45. MaxMark: High-Capacity Diffusion-Native Watermarking via Robust and Invertible Latent Embedding

46. Towards Highly Transferable Vision-Language Attack via Semantic-Augmented Dynamic Contrastive Interaction

47. FedAFD: Multimodal Federated Learning via Adversarial Fusion and Distillation

48. Learning to See Through a Baby's Eyes: Early Visual Diets Enable Robust Visual Intelligence in Humans and Machines

49. Query2Uncertainty: Robust Uncertainty Quantification and Calibration for 3D Object Detection under Distribution Shift

50. DeepProtect: Proactive Face-Swapping Defense using Identity Blending and Attribute Distortion

51. Exposing Functional Fusion: A New Class of Strategic Backdoor in Dynamic Prompt Architectures

52. Protego: User-Centric Pose-Invariant Privacy Protection Against Face Recognition-Induced Digital Footprint Exposure

53. Decoupling Defense Strategies for Robust Image Watermarking

54. What Your Features Reveal: Data-Efficient Black-Box Feature Inversion Attack for Split DNNs

55. CamPI: Physical Adversarial Examples through Camera Power Signal Injection

56. Submodel Extraction for Efficient and Personalized Federated Learning via Optimal Transport

57. ClusterMark: Towards Robust Watermarking for Autoregressive Image Generators with Visual Token Clustering

58. CoRoGS: Contextual Gaussian Splatting for Robust Large-Deviation View Synthesis

59. Look Before You Fuse: 2D-Guided Cross-Modal Alignment for Robust 3D Detection

60. Batman: Benign Knowledge Alignment Through Malicious Null Space in Federated Backdoor Attack

61. DualSplat: Robust 3D Gaussian Splatting via Pseudo-Mask Bootstrapping from Reconstruction Failures

62. TTP: Test-Time Padding for Adversarial Detection and Robust Adaptation on Vision-Language Models

63. DP-FedAdamW: An Efficient Optimizer for Differentially Private Federated Large Models

64. SAGE: Style-Adaptive Generalization for Privacy-Constrained Semantic Segmentation Across Domains

65. CASR: A Robust Cyclic Framework for Arbitrary Large-Scale Super-Resolution with Distribution Alignment and Self-Similarity Awareness

66. LDP-Slicing: Local Differential Privacy for Images via Randomized Bit-Plane Slicing

67. MERLIN: Building Low-SNR Robust Multimodal LLMs for Electromagnetic Signals

68. A Unified Perspective on Adversarial Membership Manipulation in Vision Models

69. Data-Centric Meta-Learning for Robust Few-Shot Generalization

70. CG-Reasoner: Centroid-Guided Positional Reasoning Segmentation for Medical Imaging with a Robust Visual-Text Consistency Metric

71. UAVLight: A Benchmark for Illumination-Robust 3D Reconstruction in Unmanned Aerial Vehicle (UAV) Scenes

72. RecoverMark: Robust Watermarking for Localization and Recovery of Manipulated Faces

73. ⊘\oslash Source Models Leak What They Shouldn't ↛\nrightarrow: Unlearning Zero-Shot Transfer in Domain Adaptation Through Adversarial Optimization

74. Perception Characteristics Distance: Measuring Stability and Robustness of Perception System in Dynamic Conditions under a Certain Decision Rule

75. PureProof: Diffusion-Resistant Black-box Targeted Attack on Large Vision-Language Models

76. Shedding Light on VLN Robustness: A Black-box Framework for Indoor Lighting-based Adversarial Attack

77. PAUL: Uncertainty-Guided Partition and Augmentation for Robust Cross-View Geo-Localization under Noisy Correspondence

78. FedSST: Rethinking Fair Federated Graph Learning under Structural Shift

79. UniDef: Universal Defense Against Unauthorized Image Manipulation

80. PoseGAM: Robust Unseen Object Pose Estimation via Geometry-Aware Multi-View Reasoning

81. DualMirage: Hunting Stealthy Multimodal LLM Agents via CAPTCHAs with Contour and Adversarial Illusions

82. Stabilizing Feature Geometry in Noisy Pretrained Models for Robust Downstream Tasks

83. Learning to Act Robustly with View-Invariant Latent Actions

84. A Provable Energy-Guided Test-Time Defense Boosting Adversarial Robustness of Large Vision-Language Models

85. Towards Robust Multi-Modal Semantic Segmentation with Teacher-Student Framework and Hybrid Prototype Distillation

86. Towards Robust Vision Transformers: Path Dependency Analysis and a Simple Two-Stage Adversarial Training

87. On the Role of Temporal Granularity in the Robustness of Spiking Neural Networks

88. Federated Active Learning Under Extreme Non-IID and Global Class Imbalance

89. Building Robust Vision Encoders for Cross-Dataset Evaluation in Immunofluorescent Microscopy

90. RI-Mamba: Rotation-Invariant Mamba for Robust Text-to-Shape Retrieval

91. FedMPT: Federated Multi-Label Prompt Tuning of Vision-Language Models

92. Dynamic Token Reweighting for Robust Vision-Language Models

93. Verifying Neural Network Robustness with Dual Perturbations

94. SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning

95. FedCART: Tackling Long-Tailed Distributions in Federated Adversarial Training via Classifier Refinement

96. VarSplat: Uncertainty-aware 3D Gaussian Splatting for Robust RGB-D SLAM

97. Reinforcement-Guided Synthetic Data Generation for Privacy-Sensitive Identity Recognition

98. Robust Spiking Neural Networks by Temporal Mutual Information

99. Distribution-Aligned Multimodal Fusion for Robust Object Detection

100. From Selection to Scheduling: Federated Geometry-Aware Correction Makes Exemplar Replay Work Better under Continual Dynamic Heterogeneity

101. PoInit-of-View: Poisoning Initialization of Views Transfers Across Multiple 3D Reconstruction Systems

102. Fully Decentralized Certified Unlearning

103. LAOF: Robust Latent Action Learning with Optical Flow Constraints

104. Towards Reliable Evaluation of Adversarial Robustness for Spiking Neural Networks

105. Improving Adversarial Transferability with Local Perturbation Augmentation

106. R2^2TUA: Reconstruction-residual Based Targeted and Untargeted Attack Against Text-Image Person Re-Identification

107. Hybrid Robust Collaborative Perception with LiDAR-4D Radar Fusion under Adverse Weather Conditions

108. AGFT: Alignment-Guided Fine-Tuning for Zero-Shot Adversarial Robustness of Vision-Language Models

109. PQDT: Pseudo-Query Dual Transformer for Robust Point Cloud Restoration

110. TextFM: Robust Semi-dense Feature Matching with Language Guidance

111. Concept-Guided Fine-Tuning: Steering ViTs away from Spurious Correlations to Improve Robustness

112. The Geometry of Robustness: Optimizing Loss Landscape Curvature and Feature Manifold Alignment for Robust Finetuning of Vision-Language Models

113. No Way To Steal My Face: Proactive Defense Against Identity-Preserving Personalized Generation

114. Domain Sensitive Federated Learning with Fisher-Informed Pruning

115. AdapAction: Adaptive Target Action Backdoor Attack against GUI Agents

116. Spectral Super-Resolution via Adversarial Unfolding and Data-Driven Spectrum Regularization: From Multispectral Satellite Data to NASA Hyperspectral Image

117. Generative Adversarial Perturbations with Cross-paradigm Transferability on Localized Crowd Counting

118. ProxyFL: A Proxy-Guided Framework for Federated Semi-Supervised Learning

119. FedRG: Unleashing the Representation Geometry for Federated Learning with Noisy Clients

120. ShapeR: Robust Conditional 3D Shape Generation from Casual Captures

121. Phantom: Physical Object Interactions as Dynamic Triggers for NMS-Exploited Backdoors

122. PrivSynth: Alternating and Control-Based Optimization for Privacy and Utility in Synthetic Data

123. Few-for-Many Personalized Federated Learning

124. Echoes of Ownership: Adversarial-Guided Dual Injection for Copyright Protection in MLLMs

125. Re-evaluating Continual VQA: Toward Fair and Robust Evaluation for Multimodal Continual Learning

126. GSV2X: Geometry-Aware Uncertainty Modeling and Orthogonal Fusion for Robust Roadside Perception

127. Domain-Skewed Federated Learning with Feature Decoupling and Calibration

128. When CLIP Sees More, It Fights Back Harder: Multi-View Guided Adaptive Counterattacks for Test-Time Adversarial Robustness

129. Fine-Tuning Impairs the Balancedness of Foundation Models in Long-tailed Personalized Federated Learning

130. Towards Robust Multimodal Large Language Models Against Jailbreak Attacks

131. XSeg: A Large-scale X-ray Contraband Segmentation Benchmark For Real-World Security Screening

132. SinGeo: Unlock Single Model's Potential for Robust Cross-View Geo-Localization

133. Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift

134. OmniFM: Toward Modality-Robust and Task-Agnostic Federated Learning for Heterogeneous Medical Imaging

135. FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models

136. Test-Time Attention Purification for Backdoored Large Vision Language Models

137. HiFi-BRep: High-Fidelity Latent Representation for Robust B-Rep Generation

138. IAG: Input-aware Backdoor Attack on VLM-based Visual Grounding

139. EventGait: Towards Robust Gait Recognition with Event Streams

140. Beyond Multiple Choice: Verifiable OpenQA for Robust Vision-Language RFT

141. A Cross-view Fusion Framework for Robust 6-DoF Grasp Pose Estimation

142. NEC-Diff: Noise-Robust Event-RAW Complementary Diffusion for Seeing Motion in Extreme Darkness

143. AntiStyler: Defending Object Detection Models Against Adversarial Patch Attacks Using Style Removal

144. ConeSep: Cone-based Robust Noise-Unlearning Compositional Network for Composed Image Retrieval

145. Generalized and Personalized Federated Learning with Black-Box Foundation Models via Orthogonal Transformations

146. PA-Attack: Guiding Gray-Box Attacks on LVLM Vision Encoders with Prototypes and Attention

147. DDSF: Robust Few-Shot Learning via Disentangled Subspaces with Determinantal Point Process

148. Meta-FC: Meta-Learning with Feature Consistency for Robust and Generalizable Watermarking

149. SABER: Spatially Consistent 3D Universal Adversarial Objects for BEV Detectors

150. DASH: A Meta-Attack Framework for Synthesizing Effective and Stealthy Adversarial Examples

151. Seeing Through the Shift: Causality-Inspired Robust Generalized Category Discovery

152. ComPose: A Unified Completion-Pose Framework for Robust Category-Level Object Pose Estimation

153. Robust3DGSW: Toward Robust Watermarking for Quantization-Aware 3D Gaussian Splatting

154. ULF-Loc: Unbiased Landmark Feature for Robust Visual Localization with 3D Gaussian Splatting

155. ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding

156. PECCVAI: Overcoming the Brittleness of AI Image Watermarking Under Visual Paraphrasing Attacks

157. D^3FER: Dual Channel and Dual Branch Network for Robust Facial Expression Recognition under Dual Challenges

158. When Robots Obey the Patch: Universal Transferable Patch Attacks on Vision-Language-Action Models

159. Principled Steering via Null-space Projection for Jailbreak Defense in Vision-Language Models

160. SynCLIP: Synonym-Coherent Language-Image Pretraining for Robust Open-Vocabulary Dense Perception

161. Enhancing the Security of Visual Speaker Authentication Based on Dynamic Lip-Print Analysis

162. SHARP: Short-Window Streaming for Accurate and Robust Prediction in Motion Forecasting

163. Is the Modality Gap a Bug or a Feature? A Robustness Perspective

164. Transform to Transfer: Boosting Adversarial Attack Transferability on Vision-Language Pre-training Models

165. Detecting Compressed AI-Generated Images via Phase Spectrum Robustness

166. LIBERO-Plus: A Progressive Robustness Benchmark for Visual-Language-Action Models

167. ReCoFuse: Ultra-Robust Image Fusion via Restorative Multi-Modal Diffusion Reciprocal Coupling

168. Personalized Federated Training of Diffusion Models with Privacy Guarantees

169. Rank-Guided Pseudo-Bias Learning for Robust Black-Box Adaptation

170. Pose-guided Enriched Feature Learning for Federated-by-camera Person Re-identification

171. MacTok: Robust Continuous Tokenization for Image Generation

172. CICA: Coupling Confidence-Aware Pretraining with Confidence-Informed Attention for Robust Multimodal Sentiment Analysis

173. Consistency Beyond Contrast: Enhancing Open-Vocabulary Object Detection Robustness via Contextual Consistency Learning

174. Hierarchically Robust Zero-shot Vision-language Models

175. AutoDebias: An Automated Framework for Detecting and Mitigating Backdoor Biases in Text-to-Image Models

176. ReMoE: Region-Mixture Experts for Adversarially-Robust Vision Transformers

177. DPL: Decoupled Prototype Learning for Enhancing Robustness of Vision-Language Transformers to Missing Modalities

178. AERGS-SLAM: Auto-Exposure-Robust Stereo 3D Gaussian Splatting SLAM

179. Towards Stable Federated Continual Test-Time Adaptation in Wild World

180. WaTeRFlow: Watermark Temporal Robustness via Flow Consistency

181. Your Classifier Can Do More: Towards Balancing the Gaps in Classification, Robustness, and Generation

182. Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbation

183. From Feature Learning to Spectral Basis Learning: A Unifying and Flexible Framework for Efficient and Robust Shape Matching

184. Red-teaming Retrieval-Augmented Diffusion Models via Poisoning Knowledge Bases

185. V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMs

186. Omni-Attack: Adversarial Attacks on Open-Ended VQA in Black-Box Multimodal LLMs

187. EgoXtreme: A Dataset for Robust Object Pose Estimation in Egocentric Views under Extreme Conditions

188. UniFusion: A Unified Image Fusion Framework with Robust Representation and Source-Aware Preservation

189. HFedATM: Hierarchical Federated Domain Generalization via Optimal Transport and Regularized Mean Aggregation

190. Towards Stealthy and Effective Backdoor Attacks on Lane Detection: A Naturalistic Data Poisoning Approach

191. BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation

192. Scaling Test-Time Robustness of Vision-Language Models via Self-Critical Inference Framework

193. FedRAC: Rolling Submodel Allocation for Collaborative Fairness in Federated Learning

194. Fractal Camouflage: A Bio-Inspired Approach for Multi-Scale Adversarial Attacks in the Infrared Domain

195. TANGO: Text-Anchored Guided Optimization for Robust Fine-tuning Vision-Language Models under Label Noise

196. FedRE: A Representation Entanglement Framework for Model-Heterogeneous Federated Learning

197. FedAdamom: Adaptive Momentum for Improved Generalization in Federated Optimization

198. Anchoring the Mind of Multimodal Reasoners: Cognitive Bias as a Vector for Jailbreak Attacks

199. Single-Round Scalable Analytic Federated Learning

200. FedBPrompt: Federated Domain Generalization Person Re-Identification via Body Distribution Aware Visual Prompts

201. IrisFP: Adversarial-Example-based Model Fingerprinting with Enhanced Uniqueness and Robustness

202. Unleashing Stealthy Backdoor Pandemic by Infecting a Single Diffusion Model

203. Graph Attention Prototypical Network for Robust Few-Shot Classification

204. Robust Promptable Video Object Segmentation

205. TTAPFormer: Robust Arbitrary Point Tracking via Transient Asynchronous Fusion of Frames and Events

206. RDF-MIG: A Robust Diffusion Framework for Masked Image Generation to Augment Semantic Segmentation and Change Detection

207. Robo-SGG: Exploiting Layout-Oriented Normalization and Restitution Can Improve Robust Scene Graph Generation

208. FedMOP: Achieving Enhanced Privacy and Performance in Federated Learning via Momentum Orthogonal Projection

209. Robustness Under Data Scarcity: Few-Shot Continual Adversarial Training for Evolving Threats

210. Factorized Context Aggregation for Robust Cancer Risk Estimation via Soft Re-Ranked Retrieval and Hierarchical Anchors

211. Denoise and Align: Towards Source-Free UDA for Robust Panoramic Semantic Segmentation

212. TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models

213. Tutor-Student Reinforcement Learning: A Dynamic Curriculum for Robust Deepfake Detection

214. Enhance-then-Balance Modality Collaboration for Robust Multimodal Sentiment Analysis

215. From 2D Alignment to 3D Plausibility: Unifying Heterogeneous 2D Priors and Penetration-Free Diffusion for Occlusion-Robust Two-Hand Reconstruction

216. RobustVisRAG: Causality-Aware Vision-Based Retrieval-Augmented Generation under Visual Degradations

217. RoSAMDepth: Robust Self-supervised Depth Estimation Leveraging Segment Anything Model

218. FedAlign: Differentially Private Distribution Alignment for Non-IID Federated Learning

219. AdvFM: Lookahead Flow-Matching Velocity-Field Attacks for Imperceptible and Transferable Adversarial Examples

220. Occlusion-Aware SORT: Observing Occlusion for Robust Multi-Object Tracking

221. TIACam: Text-Anchored Invariant Feature Learning with Auto-Augmentation for Camera-Robust Zero-Watermarking

222. Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations

223. VCP-Attack: Visual-Contrastive Projection for Transferable Black-Box Targeted Attacks on Large Vision-Language Models

224. Guiding Diffusion Models with Fine-Grained Conditions and Semantics-Preserving Sampling for One-Shot Federated Learning

225. WildPose: A Unified Framework for Robust Pose Estimation in the Wild

226. DSERT-RoLL: Robust Multi-Modal Perception for Diverse Driving Conditions with Stereo Event-RGB-Thermal Cameras, 4D Radar, and Dual-LiDAR

227. Personalized Longitudinal Medical Report Generation via Temporally-Aware Federated Adaptation

228. Multi-Paradigm Collaborative Adversarial Attack Against Multi-Modal Large Language Models

229. SafeLogo: Turning Your Logos into Jailbreak Shields via Micro-Regional Adversarial Training

230. VIRO: Robust and Efficient Neuro-Symbolic Reasoning with Verification for Referring Expression Comprehension

231. Bridging Privacy and Provenance: Traceable Virtual Identity Generation

232. Large-scale Robust Enhanced Ensemble Clustering via Outlier Decoupling

233. Logit-Margin Repulsion for Backdoor Defense

234. Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning

235. Learning Mutual View Information Graph for Adaptive Adversarial Collaborative Perception

236. Controllable Federated Prompt Learning at Test Time

237. Degradation-Robust Fusion: An Efficient Degradation-Aware Diffusion Framework for Multimodal Image Fusion in Arbitrary Degradation Scenarios

238. HiLoRA: Hierarchical Low-Rank Adaptation for Personalized Federated Learning

239. FreqEdit: Preserving High-Frequency Features for Robust Multi-Turn Image Editing

240. Thermally Activated Dual-Modal Adversarial Clothing against AI Surveillance Systems

241. Outlier-Robust Diffusion Solvers for Inverse Problems

242. RHO: Robust Holistic OSM-Based Metric Cross-View Geo-Localization

243. ChartR: Evaluating Reasoning Accuracy and Robustness in Chart Question Answering

244. RemedyGS: Defend 3D Gaussian Splatting Against Computation Cost Attacks

245. Good Can Sometimes be Bad: A Unified Attack against 3D Point Cloud Classifier by a Flexible Isotropic Resampling

246. The Power of Decaying Steps: Enhancing Attack Stability and Transferability for Sign-based Optimizers

247. Towards Robust Sequential Decomposition for Complex Image Editing

248. Hierarchical Attacks for Multi-Modal Multi-Agent Reasoning