VerTox is a framework that turns corpus poisoning of neural ranking models into a verifiable reward‑guided reinforcement learning problem. By fine‑tuning compact large language models with reward shaping that couples ranking distortion and factual corruption, VerTox generates fluent, low‑perplexity adversarial documents that frequently outrank target items across multiple ranking architectures, including a commercial embedding model. Experiments show near‑perfect attack success and significant degradation of downstream retrieval‑augmented generation performance.
By Zhiqi Huang, Vivek Datla, Zhichao Xu, Puxuan Yu, Vivek Srikumar, Alfy Samuel
RAGSentinel is a training‑free, label‑free defense designed for black‑box retrieval‑augmented generation systems. It employs a surrogate encoder to detect hidden‑state shifts caused by retrieved documents, removes shared topic directions, and filters poisoned documents as geometric outliers from a robust majority consensus. The method is proven to recover a poison‑free majority context under honest‑majority and representation‑separation assumptions, and experiments show it keeps attack success rates low while maintaining accuracy across multiple datasets, LLM families, and adaptive attacks.
By Yueyang Quan, Anjun Gao, Yufei Xia, Minghong Fang, Zhuqing Liu
As state-of-the-art machine translation models saturate standard benchmarks, the field needs more challenging evaluations to distinguish between models of varying quality. We propose augmenting existi...
The paper introduces MOSAIC, a large adversarial benchmark for detecting AI-generated text, and presents NeuroStat, a new framework that combines token‑level probabilistic logits with deep semantic hidden states from a single language model. NeuroStat fuses these signals via Macro‑State Residual Modulation and uses orthogonal and contrastive losses to learn complementary representations. Experiments show that NeuroStat outperforms existing methods on MOSAIC, achieving superior robustness against adversarial attacks.
By Peiming Li, Yifan Wang, Zhiyuan Hu, Shiyu Li, Zheng Wei, Yang Tang
arXiv:2608. 15932v1 Announce Type: new Abstract: As state-of-the-art machine translation models saturate standard benchmarks, the field needs more challenging evaluations to distinguish between models of varying quality.
By William Kalikman, \v{S}imon Sukup, Michal Te\v{s}nar, Vil\'em Zouhar
arXiv:2505. 19840v3 Announce Type: replace-cross Abstract: Deep Neural Networks (DNNs) have achieved widespread success yet remain prone to adversarial attacks.
By Binyan Xu, Xilin Dai, Di Tang, Kehuan Zhang
arXiv:2607. 13565v1 Announce Type: cross Abstract: We investigate which language model evasion attacks survive state-of-the-art adversarial fine-tuning, developing strategies that sweep the top 5 positions on the ELOQUENT 2026 Voight-Kampff leaderboard.
By Dima Galat, Marian-Andrei Rizoiu
arXiv:2601.03981v3 Announce Type: replace
Abstract: To efficiently combat the spread of LLM-generated misinformation in the news domain, we present RADAR, a Retrieval-Augmented Detector with Adversar...
By Song-Duo Ma, Yi-Hung Liu, Hsin-Yu Lin, Pin-Yu Chen, Hong-Yan Huang, Shau-Yung Hsu, Yun-Nung Chen
arXiv:2608. 05430v1 Announce Type: cross Abstract: The remarkable instruction-following ability of modern LLMs has enabled their practical use as the minds of agents that can autonomously complete increasingly complex tasks.
By Buzhao Liu, Xinhang Ma, Yevgeniy Vorobeychik
arXiv:2607. 20472v1 Announce Type: new Abstract: When a user asks a language model something harmful, is it a genuine attack or a misunderstood but well-meaning question?
By Roman Belaire, Arunesh Sinha, Pradeep Varakantham
arXiv:2606. 02643v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG)-enhanced LLM systems, while powerful, introduce substantial inference costs due to the inclusion of an extra multi-stage pipeline that dynamically retrieves and synthesizes information from external knowledge sources.
By Chengliang Liu, Liangbo Ning, Yujuan Ding, Wenqi Fan
arXiv:2601. 14300v4 Announce Type: replace Abstract: Hard-label black-box attacks, relying solely on top-1 predictions, represent one of the most challenging yet practically threat models.
By Jun Liu, Leo Yu Zhang, Fengpeng Li, Isao Echizen, Jiantao Zhou