NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
341 results for “keyword:"attacks"” in ICLR 2027. 100 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences▼1.2k1.67
[PDF]CoER: Defending against Adaptive Indirect Prompt Injection via Adversarial Co-Evolution and Refinement▼4.5k2.67
[PDF]When Reasoning Fails to Stop: Prompt-Induced Resource Exhaustion in Black-Box Reasoning Models▼1601.50
[PDF]Seeing Safety Across Languages: Disentangling Multilingual Safety from Visual-Language Failures in Multimodal Large Language Models▼1351.33
[PDF]MARP: Task-Generalized Meta-Attack for Promotion in CLIP-Based Text-to-Image Retrieval▲3.0k2.33
[PDF]ATFeaturePure: Learning A Class-Conditional Vector Field in Decision-Relevant Feature Space for Adversarial Defense▼1351.33
[PDF]MatBak: Targeted-Behavior Adversarial Attacks on Multiple Reinforcement Agents via Preference-Based Intention Policies▼1.2k1.67
[PDF]Bending Early, Curbing Deep: Test-Time Adversarial Defense for Large Vision-Language Models Based on Residual Curvature▼5.8k2.33
[PDF]AliasGuard: Surrogate-Based Defense Against Passive Inference in Split Language Model Fine-Tuning▼4.3k2.00
[PDF]Implicit Ranking Leakage in Black-Box RAG: Retrieval Stability-Guided Knowledge Corruption Attacks▲3.8k2.50
[PDF]When the Editing Intent Is Split: A Cross-Modal Jailbreak Attack for Large Image Editing Models▲4012.67
[PDF]Black-Box Membership Inference Attack for Vision-Language Models Using Self-Memory Probing▼1.2k1.67
Withdrawn or desk rejected (100)
[PDF]Attested but Unaccounted: DP-SGD Privacy Accounting as a Precondition of Every StepWithdrawn1.67
[PDF]TypoShift: Open-Set Retrieval Hijacking and Attention Pathway Surgery for Text-to-Image Person Re-IdentificationWithdrawn2.00
[PDF]HAFT: Adversarial Robustness Benchmarking for Hindi Fact-Checking with Adaptive Attack SelectionDesk rejected
[PDF]It Takes Two to Tango: Backdoor Attack for Multimodal Diffusion Models via Cross-modal Collaborative TriggersWithdrawn1.67
[PDF]WMAttack: Automated Attack Search for Adversarial Evaluation of World-Model AgentsWithdrawn1.67