相关论文: Hierarchical Alignment: Enforcing Hierarchical Ins…

As large language model (LLM) based systems take on high-stakes roles in real-world decision-making, they must reconcile competing instructions from multiple sources (e.g., model developers, users, and tools) within a single prompt context.…

The instruction hierarchy, which establishes a priority order from system messages to user messages, conversation history, and tool outputs, is essential for ensuring consistent and safe behavior in language models (LMs). Despite its…

Large language models (LLMs) are increasingly deployed with hierarchical instruction schemes, where certain instructions (e.g., system-level directives) are expected to take precedence over others (e.g., user messages). Yet, we lack a…

A fundamental problem of applying Large Language Models (LLMs) to important applications is that LLMs do not always follow instructions, and violations are often hard to observe or check. In LLM-based agentic workflows, such violations can…

Task semantics can be expressed by a set of input-output examples or a piece of textual instruction. Conventional machine learning approaches for natural language processing (NLP) mainly rely on the availability of large-scale sets of…

As Large Language Models (LLMs) become more powerful and autonomous, they increasingly face conflicts and dilemmas in many scenarios. We first summarize and taxonomize these diverse conflicts. Then, we model the LLM's preferences to make…

Instruction following aims to align Large Language Models (LLMs) with human intent by specifying explicit constraints on how tasks should be performed. However, we reveal a counterintuitive phenomenon: instruction following can…

Today's LLMs are susceptible to prompt injections, jailbreaks, and other attacks that allow adversaries to overwrite a model's original instructions with their own malicious prompts. In this work, we argue that one of the primary…

Large language models (LLMs) have shown tremendous success in following user instructions and generating helpful responses. Nevertheless, their robustness is still far from optimal, as they may generate significantly inconsistent responses…

Generative AI, especially via Large Language Models (LLMs), has transformed content creation across text, images, and music, showcasing capabilities in following instructions through prompting, largely facilitated by instruction tuning.…

Large language model agents receive instructions from many sources-system messages, user prompts, tool outputs, other agents, and more-each carrying different levels of trust and authority. When these instructions conflict, agents must…

Large language models (LLMs) that are tuned with instructions have demonstrated remarkable capabilities in various tasks and languages. However, their ability to generalize to underrepresented languages is limited due to the scarcity of…

Large language models should follow hierarchical instructions where system prompts override user inputs, yet recent work shows they often ignore this rule while strongly obeying social cues such as authority or consensus. We extend these…

We propose an efficient interpretable neuro-symbolic model to solve Inductive Logic Programming (ILP) problems. In this model, which is built from a set of meta-rules organised in a hierarchical structure, first-order rules are invented by…

Large Language Models (LLMs), when used in educational settings without pedagogical fine-tuning, often provide immediate answers rather than guiding students through the problem-solving process. This approach falls short of pedagogically…

Large language models (LLMs) are a promising venue for natural language understanding and generation. However, current LLMs are far from reliable: they are prone to generating non-factual information and, more crucially, to contradicting…

Hierarchical Instruction Following (HIF) refers to the problem of prompting large language models with a priority-ordered stack of instructions. Standard methods like RLHF and DPO typically fail in this problem since they mainly optimize…

Recent studies on the safety alignment of large language models (LLMs) have revealed that existing approaches often operate superficially, leaving models vulnerable to various adversarial attacks. Despite their significance, these studies…

In high-stakes scenarios-such as self-harm, legal, or medical queries-LLMs must be both trustworthy and helpful. However, these goals often conflict. We propose priority alignment, a new alignment paradigm that enforces a strict…

Large Language Models (LLMs) are susceptible to security and safety threats, such as prompt injection, prompt extraction, and harmful requests. One major cause of these vulnerabilities is the lack of an instruction hierarchy. Modern LLM…