Large language models (LLMs) have achieved remarkable progress in solving various natural language processing tasks due to emergent reasoning abilities. However, LLMs have inherent limitations as they are incapable of accessing up-to-date information (stored on the Web or in task-specific knowledge bases), using external tools, and performing precise mathematical and logical reasoning. In this paper, we present Chameleon, an AI system that mitigates these limitations by augmenting LLMs with plug-and-play modules for compositional reasoning. Chameleon synthesizes programs by composing various tools (e.g., LLMs, off-the-shelf vision models, web search engines, Python functions, and heuristic-based modules) for accomplishing complex reasoning tasks. At the heart of Chameleon is an LLM-based planner that assembles a sequence of tools to execute to generate the final response. We showcase the effectiveness of Chameleon on two multi-modal knowledge-intensive reasoning tasks: ScienceQA and TabMWP. Chameleon, powered by GPT-4, achieves an 86.54% overall accuracy on ScienceQA, improving the best published few-shot result by 11.37%. On TabMWP, GPT-4-powered Chameleon improves the accuracy by 17.0%, lifting the state of the art to 98.78%. Our analysis also shows that the GPT-4-powered planner exhibits more consistent and rational tool selection via inferring potential constraints from instructions, compared to a ChatGPT-powered planner. The project is available at https://chameleon-llm.github.io.

A long-standing goal of AI systems is to perform complex multimodal reasoning like humans. Recently, large language models (LLMs) have made remarkable strides in such multi-step reasoning on the language modality solely by leveraging the chain of thought (CoT) to mimic human thinking. However, the transfer of these advancements to multimodal contexts introduces heightened challenges, including but not limited to the impractical need for labor-intensive annotation and the limitations in terms of flexibility, generalizability, and explainability. To evoke CoT reasoning in multimodality, this work first conducts an in-depth analysis of these challenges posed by multimodality and presents two key insights: "keeping critical thinking" and "letting everyone do their jobs" in multimodal CoT reasoning. Furthermore, this study proposes a novel DDCoT prompting that maintains a critical attitude through negative-space prompting and incorporates multimodality into reasoning by first dividing the reasoning responsibility of LLMs into reasoning and recognition and then integrating the visual recognition capability of visual models into the joint reasoning process. The rationales generated by DDCoT not only improve the reasoning abilities of both large and small language models in zero-shot prompting and fine-tuning learning, significantly outperforming state-of-the-art methods but also exhibit impressive generalizability and explainability.

Open Information Extraction (OpenIE) is a fundamental yet challenging task in Natural Language Processing, which involves extracting all triples (subject, predicate, object) from a given sentence. While labeling-based methods have their merits, generation-based techniques offer unique advantages, such as the ability to generate tokens not present in the original sentence. However, these generation-based methods often require a significant amount of training data to learn the task form of OpenIE and substantial training time to overcome slow model convergence due to the order penalty. In this paper, we introduce a novel framework, OK-IE, that ingeniously transforms the task form of OpenIE into the pre-training task form of the T5 model, thereby reducing the need for extensive training data. Furthermore, we introduce an innovative concept of Anchor to control the sequence of model outputs, effectively eliminating the impact of order penalty on model convergence and significantly reducing training time. Experimental results indicate that, compared to previous SOTA methods, OK-IE requires only 1/100 of the training data (900 instances) and 1/120 of the training time (3 minutes) to achieve comparable results.

Evaluating environmental variables that vary stochastically is the principal topic for designing better environmental management and restoration schemes. Both the upper and lower estimates of these variables, such as water quality indices and flood and drought water levels, are important and should be consistently evaluated within a unified mathematical framework. We propose a novel pair of Orlicz regrets to consistently bound the statistics of random variables both from below and above. Here, consistency indicates that the upper and lower bounds are evaluated with common coefficients and parameter values being different from some of the risk measures proposed thus far. Orlicz regrets can flexibly evaluate the statistics of random variables based on their tail behavior. The explicit linkage between Orlicz regrets and divergence risk measures was exploited to better comprehend them. We obtain sufficient conditions to pose the Orlicz regrets as well as divergence risk measures, and further provide gradient descent-type numerical algorithms to compute them. Finally, we apply the proposed mathematical framework to the statistical evaluation of 31-year water quality data as key environmental indicators in a Japanese river environment.

The warming of the Arctic, also known as Arctic amplification, is led by several atmospheric and oceanic drivers. However, the details of its underlying thermodynamic causes are still unknown. Inferring the causal effects of atmospheric processes on sea ice melt using fixed treatment effect strategies leads to unrealistic counterfactual estimations. Such models are also prone to bias due to time-varying confoundedness. Further, the complex non-linearity in Earth science data makes it infeasible to perform causal inference using existing marginal structural techniques. In order to tackle these challenges, we propose TCINet - time-series causal inference model to infer causation under continuous treatment using recurrent neural networks and a novel probabilistic balancing technique. Through experiments on synthetic and observational data, we show how our research can substantially improve the ability to quantify leading causes of Arctic sea ice melt, further paving paths for causal inference in observational Earth science.

Establishing a human settlement on Mars is an incredibly complex engineering problem. The inhospitable nature of the Martian environment requires any habitat to be largely self-sustaining. Beyond mining a few basic minerals and water, the colonizers will be dependent on Earth resupply and replenishment of necessities via technological means, i.e., splitting Martian water into oxygen for breathing and hydrogen for fuel. Beyond the technical and engineering challenges, future colonists will also face psychological and human behavior challenges. Our goal is to better understand the behavioral and psychological interactions of future Martian colonists through an Agent-Based Modeling (ABM simulation) approach. We seek to identify areas of consideration for planning a colony as well as propose a minimum initial population size required to create a stable colony. Accounting for engineering and technological limitations, we draw on research regarding high performing teams in isolated and high stress environments (ex: submarines, Arctic exploration, ISS, war) to include the 4 basic personality types within the ABM. Interactions between agents with different psychological profiles are modeled at the individual level, while global events such as accidents or delays in Earth resupply affect the colony as a whole. From our multiple simulations and scenarios (up to 28 Earth years), we found that an initial population of 22 was the minimum required to maintain a viable colony size over the long run. We also found that the agreeable personality type was the one more likely to survive. We find, contrary to other literature, that the minimum number of people with all personality types that can lead to a sustainable settlement is in the tens and not hundreds.

The accurate assessment of fish stocks is crucial for sustainable fisheries management. However, existing statistical stock assessment models can have low forecast performance of relevant stock parameters like recruitment or spawning stock biomass, especially in ecosystems that are changing due to global warming and other anthropogenic stressors. In this paper, we investigate the use of machine learning models to improve the estimation and forecast of such stock parameters. We propose a hybrid model that combines classical statistical stock assessment models with supervised ML, specifically gradient boosted trees. Our hybrid model leverages the initial estimate provided by the classical model and uses the ML model to make a post-hoc correction to improve accuracy. We experiment with five different stocks and find that the forecast accuracy of recruitment and spawning stock biomass improves considerably in most cases.

This chapter introduces the concept of adversarial attacks on image classification models built on convolutional neural networks (CNN). CNNs are very popular deep-learning models which are used in image classification tasks. However, very powerful and pre-trained CNN models working very accurately on image datasets for image classification tasks may perform disastrously when the networks are under adversarial attacks. In this work, two very well-known adversarial attacks are discussed and their impact on the performance of image classifiers is analyzed. These two adversarial attacks are the fast gradient sign method (FGSM) and adversarial patch attack. These attacks are launched on three powerful pre-trained image classifier architectures, ResNet-34, GoogleNet, and DenseNet-161. The classification accuracy of the models in the absence and presence of the two attacks are computed on images from the publicly accessible ImageNet dataset. The results are analyzed to evaluate the impact of the attacks on the image classification task.

The goal of this paper is open-vocabulary object detection (OVOD) $\unicode{x2013}$ building a model that can detect objects beyond the set of categories seen at training, thus enabling the user to specify categories of interest at inference without the need for model retraining. We adopt a standard two-stage object detector architecture, and explore three ways for specifying novel categories: via language descriptions, via image exemplars, or via a combination of the two. We make three contributions: first, we prompt a large language model (LLM) to generate informative language descriptions for object classes, and construct powerful text-based classifiers; second, we employ a visual aggregator on image exemplars that can ingest any number of images as input, forming vision-based classifiers; and third, we provide a simple method to fuse information from language descriptions and image exemplars, yielding a multi-modal classifier. When evaluating on the challenging LVIS open-vocabulary benchmark we demonstrate that: (i) our text-based classifiers outperform all previous OVOD works; (ii) our vision-based classifiers perform as well as text-based classifiers in prior work; (iii) using multi-modal classifiers perform better than either modality alone; and finally, (iv) our text-based and multi-modal classifiers yield better performance than a fully-supervised detector.

It has always been an important yet challenging problem to control language models to avoid generating texts with undesirable attributes, such as toxic language and unnatural repetition. We introduce Click for controllable text generation, which needs no modification to the model architecture and facilitates out-of-the-box use of trained models. It employs a contrastive loss on sequence likelihood, which fundamentally decreases the generation probability of negative samples (i.e., generations with undesirable attributes). It also adopts a novel likelihood ranking-based strategy to construct contrastive samples from model generations. On the tasks of language detoxification, sentiment steering, and repetition reduction, we show that Click outperforms strong baselines of controllable text generation and demonstrate the superiority of Click's sample construction strategy.