Related papers: Estimating Tail Risks in Language Model Output Dis…

Large language models (LLMs) have demonstrated immense utility across various industries. However, as LLMs advance, the risk of harmful outputs increases due to incorrect or malicious instruction prompts. While current methods effectively…

In this paper we consider the problem of computing tail probabilities of the distribution of a random sum of positive random variables. We assume that the individual variables follow a reproducible natural exponential family (NEF)…

While large language models (LLMs) exhibit remarkable capabilities across a wide range of tasks, they pose potential safety concerns, such as the ``jailbreak'' problem, wherein malicious instructions can manipulate LLMs to exhibit…

We investigate a way of comparing and classifying tails of random variables. Our approach extends the notion of classical indices, such as exponential and moment indices, which are widely used measuring heaviness of tail functions. A…

As Multimodal Large Language Models (MLLMs) become an indispensable assistant in human life, the unsafe content generated by MLLMs poses a danger to human behavior, perpetually overhanging human society like a sword of Damocles. To…

We study the stochastic multi-armed bandit problem and design new policies that enjoy both worst-case optimality for expected regret and light-tailed risk for regret distribution. Specifically, our policy design (i) enjoys the worst-case…

Large language models (LLMs) are trained on web-scale corpora that exhibit steep power-law distributions, in which the distribution of knowledge is highly long-tailed, with most appearing infrequently. While scaling has improved…

Meta learning is a promising paradigm in the era of large models and task distributional robustness has become an indispensable consideration in real-world scenarios. Recent advances have examined the effectiveness of tail task risk…

This paper presents a systematic evaluation of Large Language Models' (LLMs) behavior on long-tail distributed (encrypted) texts and their safety implications. We introduce a two-dimensional framework for assessing LLM safety: (1)…

Standard statistical analysis is unable to provide reliable confidence intervals on expectation values of probability distributions that do not satisfy the conditions of the central limit theorem. We present a regression-based estimator of…

Large audio-language models (LALMs) extend text-based LLMs with auditory understanding, offering new opportunities for multimodal applications. While their perception, reasoning, and task performance have been widely studied, their safety…

With the rapid popularity of large language models such as ChatGPT and GPT-4, a growing amount of attention is paid to their safety concerns. These models may generate insulting and discriminatory content, reflect incorrect social values,…

Value-at-Risk (VaR) estimation at high confidence levels is inherently a rare-event problem and is particularly sensitive to tail behavior and model misspecification. This paper studies the performance of two simulation-based VaR estimation…

Identifying risk spillovers in financial markets is of great importance for assessing systemic risk and portfolio management. Granger causality in tail (or in risk) tests whether past extreme events of a time series help predicting future…

Fine-tuning language models is commonly believed to inevitably harm their safety, i.e., refusing to respond to harmful user requests, even when using harmless datasets, thus requiring additional safety measures. We challenge this belief…

In this paper we propose a problem-driven scenario generation approach to the single-period portfolio selection problem which use tail risk measures such as conditional value-at-risk. Tail risk measures are useful for quantifying potential…

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…

As Multimodal Large Language Models (MLLMs) acquire stronger reasoning capabilities to handle complex, multi-image instructions, this advancement may pose new safety risks. We study this problem by introducing MIR-SafetyBench, the first…

This paper attempts to provide a decision-theoretic foundation for the measurement of economic tail risk, which is not only closely related to utility theory but also relevant to statistical model uncertainty. The main result is that the…

This paper provides an introductory overview of how one may employ importance sampling effectively as a tool for solving stochastic optimization formulations incorporating tail risk measures such as Conditional Value-at-Risk. Approximating…