Search: “red-teaming”

arXiv:2607.26314v1 Announce Type: cross Abstract: Stealth, the discipline of achieving an objective without revealing your presence, capabilities, or collected intelligence, is what separates sophisti

arXiv:2607.26998v1 Announce Type: cross Abstract: Large language model (LLM) agents automate penetration testing through an observation-action loop, selecting actions based on observations returned by

arXiv:2607.26060v1 Announce Type: new Abstract: LLM-based chatbots are transforming customer service in regulated domains such as banking, but scalable and cost-effective validation remains a critical

arXiv:2607.24162v1 Announce Type: new Abstract: Optimizing agentic workflows, such as retrieval-augmented generation (RAG) pipelines, requires navigating a combinatorial space of discrete component ch

arXiv:2607.23496v1 Announce Type: new Abstract: Safety-aligned large language models are trained to refuse harmful requests, yet embedding the same requests in particular scenarios can bypass their sa

arXiv:2607.24339v1 Announce Type: new Abstract: Large language model (LLM) agents inherit reactive failure modes: escalation under provocation, sycophantic drift under flattery, perseveration when stu

arXiv:2607.22643v1 Announce Type: new Abstract: Multimodal retrieval-augmented generation (mRAG) aims to answer image-text queries with external knowledge, but most existing systems still retrieve dir

arXiv:2603.03536v2 Announce Type: replace-cross Abstract: Current LLM-based conversational recommender systems (CRS) primarily optimize recommendation accuracy and user satisfaction. We identify an un

arXiv:2510.12993v3 Announce Type: replace Abstract: Large Language Models (LLMs) can generate highly persuasive disinformation, yet little is known about how effectively they personalise it across lan

arXiv:2607.21619v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have achieved impressive performance, but their safety alignment remains vulnerable to jailbreak attacks. Exist

arXiv:2607.21758v1 Announce Type: new Abstract: Teachers, conference chairs, and public readers all judge writing from limited evidence, seeing only a finished document and not the process that produc

arXiv:2607.18292v2 Announce Type: replace-cross Abstract: As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability. The knowledge-gap account --

arXiv:2603.13026v2 Announce Type: replace Abstract: Prompt injection poses serious security risks to real-world LLM applications, particularly autonomous agents. Although many defenses have been propo

arXiv:2511.05865v3 Announce Type: replace-cross Abstract: Recent advancements in large-scale generative models have enabled the creation of high-quality images and videos, but have also raised signifi

arXiv:2607.13394v1 Announce Type: cross Abstract: Generative Flow Networks (GFlowNets) offer a promising alternative to reward-maximizing reinforcement learning (RL) for large reasoning models, encour

arXiv:2607.12894v1 Announce Type: new Abstract: Building capable embodied agents requires not only multimodal perception and understanding, but also agentic capabilities for reasoning about actions, a

arXiv:2601.11443v3 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) has emerged as a powerful approach for enhancing large language models' question-answering capabilities through

arXiv:2607.08652v1 Announce Type: new Abstract: Self-interested agents, left unconstrained, tend toward defection in repeated social dilemmas, causing cooperative gains from trade to collapse. This pa

arXiv:2607.02079v1 Announce Type: new Abstract: We present HaloGuard 1.0, an open-weights implementation of the constitutional-classifier paradigm for input safety. It achieves state-of-the-art perfor

arXiv:2607.00481v1 Announce Type: cross Abstract: Jailbreak attacks remain a critical threat to the safe deployment of large language models (LLMs). While prior work has primarily studied attacks and

arXiv:2606.29801v1 Announce Type: new Abstract: Text-to-image diffusion models remain vulnerable to adversarial prompts that elicit disallowed content, motivating reliable inference-time controls. A p

arXiv:2606.27091v1 Announce Type: cross Abstract: LLMs fine-tuned for security classification are usually evaluated on held-out examples from the same distribution as their training data. We show that

arXiv:2606.23496v1 Announce Type: new Abstract: Discrete text-trigger optimization -- searching for text sequences that, when ingested by a model, steer it toward a specified objective -- underpins mo

arXiv:2606.09556v1 Announce Type: new Abstract: AI Scientist agents are often evaluated as if capability were mainly a function of model quality, prompting, or reasoning scaffolds. We test a different

arXiv:2510.17947v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are improving at an exceptional rate. With the advent of agentic workflows, multi-turn dialogue has become the de

arXiv:2603.22793v2 Announce Type: replace Abstract: Classroom AI systems increasingly infer high-level educational states such as engagement, confusion, collaboration, participation, and instructional

arXiv:2606.06529v1 Announce Type: new Abstract: An attacker that strategically chooses when to attack is much harder to catch than one that attacks indiscriminately. AI control is a safety framework f

arXiv:2606.04027v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) generate text by iteratively denoising partially masked sequences under bidirectional context, exposing a safe

arXiv:2606.01695v1 Announce Type: new Abstract: Adversaries can implant latent harmful behavior by poisoning as few as 1% of fine-tuning examples. The contamination is invisible to every output-level

arXiv:2606.00813v1 Announce Type: cross Abstract: Safety alignment in LLMs does not improve monotonically across model generations. Studying four generations of Google's Gemma family (7B-31B) with qua

arXiv:2512.20806v3 Announce Type: replace Abstract: Ensuring the safety of language models (LMs) while maintaining their usefulness remains a critical challenge in AI alignment. Current approaches rel

arXiv:2605.31593v1 Announce Type: cross Abstract: Language models can find thousands of severe software vulnerabilities, and agents are increasingly being misused for cyberattacks. To avoid detection,

arXiv:2601.03134v2 Announce Type: replace Abstract: As LLMs gain persuasive capabilities through extended dialogues, they create new opportunities for studying adversarial conversational behavior in e

arXiv:2605.29708v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) LLMs rely on sparse, router-driven expert activation, yet how safety alignment interacts with routed expert specialization rema

arXiv:2605.28609v1 Announce Type: new Abstract: Forensic vision-language models (VLMs) have recently been developed to detect image tampering and provide natural-language explanations. However, their

arXiv:2605.26954v1 Announce Type: new Abstract: Safety evaluation of Large Language Models (LLMs) has largely focused on high-resource languages, leaving low-resource languages critically underserved.

arXiv:2510.17759v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) extend large language models with visual reasoning, but their multimodal design also introduces new, underexplor

arXiv:2605.24883v1 Announce Type: new Abstract: The widespread integration of Large Language Models (LLMs) necessitates rigorous and systematic safety evaluation. Existing paradigms either rely on con

arXiv:2605.23157v1 Announce Type: new Abstract: The attack surface of a multimodal large language model (MLLM) is language-dependent in ways that reveal the mechanistic structure of alignment failures

arXiv:2603.14987v2 Announce Type: replace Abstract: Agentic AI systems increasingly act through tool-augmented, multi-step workflows whose failures (unsafe tool use, unauthorised actions, social harm)

arXiv:2605.21362v1 Announce Type: new Abstract: Jailbreak attacks expose a persistent gap between the intended safety behavior of aligned large language models and their behavior under adversarial pro

arXiv:2605.19190v1 Announce Type: cross Abstract: Despite the global deployment of text-to-image (T2I) models, their safety frameworks are largely calibrated to a Western-centric default, creating sig

arXiv:2605.19192v1 Announce Type: new Abstract: Multimodal agents use screenshots, documents, and webpages to choose tool calls. When a false visual claim triggers a click, email, extraction, or trans

arXiv:2605.17971v1 Announce Type: cross Abstract: Despite rigorous safety alignment, Large Language Models (LLMs) remain vulnerable to jailbreak attacks. Existing black-box methods often rely on heuri

arXiv:2503.02574v2 Announce Type: replace-cross Abstract: In this paper, we argue that current safety alignment research efforts for large language models are hindered by many intertwined sources of n

arXiv:2505.11708v3 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) agents are increasingly used to simulate sophisticated cyberattacks, but their decision-making processes remain op