Related papers: Peak + Accumulation: A Proxy-Level Scoring Formula…

Large language models (LLMs) have acquired the ability to handle longer context lengths and understand nuances in text, expanding their dialogue capabilities beyond a single utterance. A popular user-facing application of LLMs is the…

Reinforcement learning can train LLM agents from sparse task rewards, but long-horizon credit assignment remains challenging: a single success-or-failure signal must be distributed across many actions. Existing methods rely on…

Large Language Models (LLMs) deployed in enterprise settings (e.g., as Microsoft 365 Copilot) face novel security challenges. One critical threat is prompt inference attacks: adversaries chain together seemingly benign prompts to gradually…

While confidence estimation is a promising direction for mitigating hallucinations in Large Language Models (LLMs), current research overwhelmingly focuses on single-turn settings. The dynamics of model confidence in multi-turn…

The advent of Large Language Models LLMs marks a milestone in Artificial Intelligence, altering how machines comprehend and generate human language. However, LLMs are vulnerable to malicious prompt injection attacks, where crafted inputs…

Large Language Models (LLMs) have gained significant attention but also raised concerns due to the risk of misuse. Jailbreak prompts, a popular type of adversarial attack towards LLMs, have appeared and constantly evolved to breach the…

Deploying LLMs in multi-turn dialogues facilitates jailbreak attacks that distribute harmful intent across seemingly benign turns. Recent training-based multi-turn jailbreak methods learn long-horizon attack strategies from interaction…

With the rapid advancement of Large Language Models (LLMs), the safety of LLMs has been a critical concern requiring precise assessment. Current benchmarks primarily concentrate on single-turn dialogues or a single jailbreak attack method…

Large language model (LLM) safety classifiers such as Llama Guard are effective at detecting overtly harmful prompts but remain vulnerable to adversarial jailbreak attacks that disguise malicious intent through role-play scenarios,…

Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources. Large Language Models (LLMs) struggle to perform such reasoning consistently. Here we propose an approach to pinpoint and rectify…

Large Language Models (LLMs) are increasingly exposed to adaptive jailbreaking, particularly in high-stakes Chemical, Biological, Radiological, and Nuclear (CBRN) domains. Although streaming probes enable real-time monitoring, they still…

Prompt Recovery, reconstructing prompts from the outputs of large language models (LLMs), has grown in importance as LLMs become ubiquitous. Most users access LLMs through APIs without internal model weights, relying only on outputs and…

With the widespread use of multi-modal Large Language models (MLLMs), safety issues have become a growing concern. Multi-turn dialogues, which are more common in everyday interactions, pose a greater risk than single prompts; however,…

Large Language Models (LLMs) remain vulnerable to jailbreaking attacks where adversarial prompts elicit harmful outputs. Yet most evaluations focus on single-turn interactions while real-world attacks unfold through adaptive multi-turn…

Large language models (LLMs) remain vulnerable to multi-turn jailbreaking attacks that exploit conversational context to bypass safety constraints gradually. These attacks target different harm categories through distinct conversational…

Multi-turn jailbreak attacks progressively erode LLM safety alignment across seemingly innocuous conversation turns, achieving success rates exceeding 90% against state-of-the-art models. Existing alignment-based and guardrail methods…

Large language models (LLMs) have shown remarkable performance across a range of NLP tasks. However, their strong instruction-following capabilities and inability to distinguish instructions from data content make them vulnerable to…

Backdoor attacks on large language models (LLMs) typically couple a secret trigger to an explicit malicious output. We show that this explicit association is unnecessary for common LLMs. We introduce a compliance-only backdoor: supervised…

Most LLM benchmarks score how well a model responds to explicit requests. They leave unmeasured a different conversational ability: noticing and acting on needs the user has implied but not said. We call this \emph{conversational…

System prompt configuration can make the difference between near-total phishing blindness and near-perfect detection in LLM email agents. We present PhishNChips, a study of 11 models under 10 prompt strategies, showing that prompt-model…