Related papers: Large-Scale Analysis of Style Injection by Relativ…

Neural ranking models (NRMs) have shown remarkable success in recent years, especially with pre-trained language models. However, deep neural models are notorious for their vulnerability to adversarial examples. Adversarial attacks may…

Web attacks, i.e. attacks exclusively using the HTTP protocol, are rapidly becoming one of the fundamental threats for information systems connected to the Internet. When the attacks suffered by web servers through the years are analyzed,…

Web-browsing AI agents are increasingly deployed in enterprise settings under strict whitelists of approved domains, yet adversaries can still influence them by embedding hidden instructions in the HTML pages those domains serve. Existing…

We present a study of how local frames (i.e., iframes loading content like "about:blank") are mishandled by a wide range of popular Web security and privacy tools. As a result, users of these tools remain vulnerable to the very attack…

SQL Injection is one of the vulnerabilities in OWASPs Top Ten List for Web Based Application Exploitation.These types of attacks takes place on Dynamic Web applications as they interact with the databases for the various operations.Current…

Vision-Language-Action (VLA) models such as OpenVLA, Octo, and $\pi_0$ have shown strong generalization by leveraging large-scale demonstrations, yet their performance is still fundamentally constrained by the quality and coverage of…

Open Radio Access Network (Open RAN) is a new paradigm to provide fundamental features for supporting next-generation mobile networks. Disaggregation, virtualisation, closed-loop data-driven control, and open interfaces bring flexibility…

Differentiable reinforcement learning (RL) frameworks like DiffRO offer a powerful approach for controllable text-to-speech (TTS), but are vulnerable to reward hacking, particularly for nuanced tasks like emotion control. The policy model…

Group Relative Policy Optimization (GRPO) has demonstrated wide adoption in the post-training of Large Language Models (LLMs). In GRPO, prompts are answered by the model and preferred behaviour is learnt via reinforcement learning. Owing to…

Automatic Prompt Optimization (APO) has emerged as a critical technique for enhancing Large Language Model (LLM) performance, yet current state-of-the-art methods typically rely on large, labeled gold-standard development sets to compute…

The Internet of Things (IoT) is becoming ubiquitous in our daily life. IoT networks that are made up of devices low power, low memory, and low computing capability appears in many applications such as healthcare, home, agriculture. IPv6…

We present a novel framework addressing a critical vulnerability in Large Language Models (LLMs): the prevalence of factual inaccuracies within intermediate reasoning steps despite correct final answers. This phenomenon poses substantial…

Adversarial attacks have proved to be the major impediment in the progress on research towards reliable machine learning solutions. Carefully crafted perturbations, imperceptible to human vision, can be added to images to force…

Group relative policy optimization (GRPO) has demonstrated significant potential in improving the reasoning capabilities of large language models (LLMs) via reinforcement learning. However, its practical deployment is impeded by an…

Optical Character Recognition (OCR) is a widely used tool to extract text from scanned documents. Today, the state-of-the-art is achieved by exploiting deep neural networks. However, the cost of this performance is paid at the price of…

Offline reinforcement learning (RL) refers to the problem of learning policies entirely from a large batch of previously collected data. This problem setting offers the promise of utilizing such datasets to acquire policies without any…

Recently, NLP has seen a surge in the usage of large pre-trained models. Users download weights of models pre-trained on large datasets, then fine-tune the weights on a task of their choice. This raises the question of whether downloading…

Existing Reinforcement Learning from Verifiable Rewards (RLVR) methods, such as Group Relative Policy Optimization (GRPO), have achieved remarkable progress in improving the reasoning capabilities of Large Reasoning Models (LRMs). However,…

The use of large language models (LLMs) in peer review systems has attracted growing attention, making it essential to examine their potential vulnerabilities. Prior attacks rely on prompt injection, which alters manuscript content and…

Recommender Systems~(RS) have been shown to be vulnerable to injective attacks, where attackers inject limited fake user profiles to promote the exposure of target items to real users for unethical gains (e.g., economic or political…