We introduce Inference-Time Intervention (ITI), a technique designed to enhance the "truthfulness" of large language models (LLMs). ITI operates by shifting model activations during inference, following a set of directions across a limited number of attention heads. This intervention significantly improves the performance of LLaMA models on the TruthfulQA benchmark. On an instruction-finetuned LLaMA called Alpaca, ITI improves its truthfulness from 32.5% to 65.1%. We identify a trade-off between truthfulness and helpfulness and demonstrate how to balance it by tuning the intervention strength. ITI is minimally invasive and computationally inexpensive. Moreover, the technique is data efficient: while approaches like RLHF require extensive annotations, ITI locates truthful directions using only few hundred examples. Our findings suggest that LLMs may have an internal representation of the likelihood of something being true, even as they produce falsehoods on the surface.

Radon is a carcinogenic, radioactive gas that can accumulate indoors. Indoor radon exposure at the national scale is usually estimated on the basis of extensive measurement campaigns. However, characteristics of the sample often differ from the characteristics of the population due to the large number of relevant factors such as the availability of geogenic radon or floor level. Furthermore, the sample size usually does not allow exposure estimation with high spatial resolution. We propose a model-based approach that allows a more realistic estimation of indoor radon distribution with a higher spatial resolution than a purely data-based approach. We applied a two-stage modelling approach: 1) a quantile regression forest using environmental and building data as predictors was applied to estimate the probability distribution function of indoor radon for each floor level of each residential building in Germany; (2) a probabilistic Monte Carlo sampling technique enabled the combination and population weighting of floor-level predictions. In this way, the uncertainty of the individual predictions is effectively propagated into the estimate of variability at the aggregated level. The results give an arithmetic mean of 63 Bq/m3, a geometric mean of 41 Bq/m3 and a 95 %ile of 180 Bq/m3. The exceedance probability for 100 Bq/m3 and 300 Bq/m3 are 12.5 % (10.5 million people) and 2.2 % (1.9 million people), respectively. In large cities, individual indoor radon exposure is generally lower than in rural areas, which is a due to the different distribution of the population on floor levels. The advantages of our approach are 1) an accurate exposure estimation even if the survey was not fully representative with respect to the main controlling factors, and 2) an estimate of the exposure distribution with a much higher spatial resolution than basic descriptive statistics.

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems. Guardrails (or rails for short) are a specific way of controlling the output of an LLM, such as not talking about topics considered harmful, following a predefined dialogue path, using a particular language style, and more. There are several mechanisms that allow LLM providers and developers to add guardrails that are embedded into a specific model at training, e.g. using model alignment. Differently, using a runtime inspired from dialogue management, NeMo Guardrails allows developers to add programmable rails to LLM applications - these are user-defined, independent of the underlying LLM, and interpretable. Our initial results show that the proposed approach can be used with several LLM providers to develop controllable and safe LLM applications using programmable rails.

In this paper, we delve into the problem of simplicial representation learning utilizing the 1-Wasserstein distance on a tree structure (a.k.a., Tree-Wasserstein distance (TWD)), where TWD is defined as the L1 distance between two tree-embedded vectors. Specifically, we consider a framework for simplicial representation estimation employing a self-supervised learning approach based on SimCLR with a negative TWD as a similarity measure. In SimCLR, the cosine similarity with real-vector embeddings is often utilized; however, it has not been well studied utilizing L1-based measures with simplicial embeddings. A key challenge is that training the L1 distance is numerically challenging and often yields unsatisfactory outcomes, and there are numerous choices for probability models. Thus, this study empirically investigates a strategy for optimizing self-supervised learning with TWD and find a stable training procedure. More specifically, we evaluate the combination of two types of TWD (total variation and ClusterTree) and several simplicial models including the softmax function, the ArcFace probability model, and simplicial embedding. Moreover, we propose a simple yet effective Jeffrey divergence-based regularization method to stabilize the optimization. Through empirical experiments on STL10, CIFAR10, CIFAR100, and SVHN, we first found that the simple combination of softmax function and TWD can obtain significantly lower results than the standard SimCLR (non-simplicial model and cosine similarity). We found that the model performance depends on the combination of TWD and the simplicial model, and the Jeffrey divergence regularization usually helps model training. Finally, we inferred that the appropriate choice of combination of TWD and simplicial models outperformed cosine similarity based representation learning.

Stealthy multi-agent active search is the problem of making efficient sequential data-collection decisions to identify an unknown number of sparsely located targets while adapting to new sensing information and concealing the search agents' location from the targets. This problem is applicable to reconnaissance tasks wherein the safety of the search agents can be compromised as the targets may be adversarial. Prior work usually focuses either on adversarial search, where the risk of revealing the agents' location to the targets is ignored or evasion strategies where efficient search is ignored. We present the Stealthy Terrain-Aware Reconnaissance (STAR) algorithm, a multi-objective parallelized Thompson sampling-based algorithm that relies on a strong topographical prior to reason over changing visibility risk over the course of the search. The STAR algorithm outperforms existing state-of-the-art multi-agent active search methods on both rate of recovery of targets as well as minimising risk even when subject to noisy observations, communication failures and an unknown number of targets.

A deep learning platform has been developed to forecast the occurrence of the low visibility events or hazes. It is trained by using multi-decadal daily regional maps of various meteorological and hydrological variables as input features and surface visibility observations as the targets. To better preserve the characteristic spatial information of different input features for training, two branched architectures have recently been developed for the case of Paris hazes. These new architectures have improved the performance of the network, producing reasonable scores in both validation and a blind forecasting evaluation using the data of 2021 and 2022 that have not been used in the training and validation.

This article presents a comprehensive overview of the digital twin technology and its capability levels, with a specific focus on its applications in the wind energy industry. It consolidates the definitions of digital twin and its capability levels on a scale from 0-5; 0-standalone, 1-descriptive, 2-diagnostic, 3-predictive, 4-prescriptive, 5-autonomous. It then, from an industrial perspective, identifies the current state of the art and research needs in the wind energy sector. The article proposes approaches to the identified challenges from the perspective of research institutes and offers a set of recommendations for diverse stakeholders to facilitate the acceptance of the technology. The contribution of this article lies in its synthesis of the current state of knowledge and its identification of future research needs and challenges from an industry perspective, ultimately providing a roadmap for future research and development in the field of digital twin and its applications in the wind energy industry.

The growing capabilities and increasingly widespread deployment of AI systems necessitate robust benchmarks for measuring their cooperative capabilities. Unfortunately, most multi-agent benchmarks are either zero-sum or purely cooperative, providing limited opportunities for such measurements. We introduce a general-sum variant of the zero-sum board game Diplomacy -- called Welfare Diplomacy -- in which players must balance investing in military conquest and domestic welfare. We argue that Welfare Diplomacy facilitates both a clearer assessment of and stronger training incentives for cooperative capabilities. Our contributions are: (1) proposing the Welfare Diplomacy rules and implementing them via an open-source Diplomacy engine; (2) constructing baseline agents using zero-shot prompted language models; and (3) conducting experiments where we find that baselines using state-of-the-art models attain high social welfare but are exploitable. Our work aims to promote societal safety by aiding researchers in developing and assessing multi-agent AI systems. Code to evaluate Welfare Diplomacy and reproduce our experiments is available at https://github.com/mukobi/welfare-diplomacy.

Planning informative trajectories while considering the spatial distribution of the information over the environment, as well as constraints such as the robot's limited battery capacity, makes the long-time horizon persistent coverage problem complex. Ergodic search methods consider the spatial distribution of environmental information while optimizing robot trajectories; however, current methods lack the ability to construct the target information spatial distribution for environments that vary stochastically across space and time. Moreover, current coverage methods dealing with battery capacity constraints either assume simple robot and battery models, or are computationally expensive. To address these problems, we propose a framework called Eclares, in which our contribution is two-fold. 1) First, we propose a method to construct the target information spatial distribution for ergodic trajectory optimization using clarity, an information measure bounded between [0,1]. The clarity dynamics allows us to capture information decay due to lack of measurements and to quantify the maximum attainable information in stochastic spatiotemporal environments. 2) Second, instead of directly tracking the ergodic trajectory, we introduce the energy-aware (eware) filter, which iteratively validates the ergodic trajectory to ensure that the robot has enough energy to return to the charging station when needed. The proposed eware filter is applicable to nonlinear robot models and is computationally lightweight. We demonstrate the working of the framework through a simulation case study.

Heavy tails is a common feature of filtering distributions that results from the nonlinear dynamical and observation processes as well as the uncertainty from physical sensors. In these settings, the Kalman filter and its ensemble version - the ensemble Kalman filter (EnKF) - that have been designed under Gaussian assumptions result in degraded performance. t-distributions are a parametric family of distributions whose tail-heaviness is modulated by a degree of freedom $\nu$. Interestingly, Cauchy and Gaussian distributions correspond to the extreme cases of a t-distribution for $\nu = 1$ and $\nu = \infty$, respectively. Leveraging tools from measure transport (Spantini et al., SIAM Review, 2022), we present a generalization of the EnKF whose prior-to-posterior update leads to exact inference for t-distributions. We demonstrate that this filter is less sensitive to outlying synthetic observations generated by the observation model for small $\nu$. Moreover, it recovers the Kalman filter for $\nu = \infty$. For nonlinear state-space models with heavy-tailed noise, we propose an algorithm to estimate the prior-to-posterior update from samples of joint forecast distribution of the states and observations. We rely on a regularized expectation-maximization (EM) algorithm to estimate the mean, scale matrix, and degree of freedom of heavy-tailed \textit{t}-distributions from limited samples (Finegold and Drton, arXiv preprint, 2014). Leveraging the conditional independence of the joint forecast distribution, we regularize the scale matrix with an $l1$ sparsity-promoting penalization of the log-likelihood at each iteration of the EM algorithm. By sequentially estimating the degree of freedom at each analysis step, our filter can adapt its prior-to-posterior update to the tail-heaviness of the data. We demonstrate the benefits of this new ensemble filter on challenging filtering problems.