自从商业领域聊天机器人出现以来，它们已被广泛地在客户服务部门工作。通常，这些商业聊天机器人是基于检索的，因此他们无法响应提供的数据集中缺少的查询。相反，生成性聊天机器人尝试创建最合适的响应，但大多无法在客户机器人对话框中创建平稳的流程。由于客户在收到响应后仍有几乎没有选择的选择，因此对话框变得短缺。通过我们的工作，我们试图最大程度地提高简单的对话代理的智能，以便可以回答看不见的查询，并产生后续问题或备注。我们已经为一家珠宝店建造了聊天机器人，该聊天机器人通过找到与语料库中模式的输入相似，从而找到了客户查询的根本目标。我们的系统为客户提供了音频输入界面，因此他们可以用自然语言对其进行交谈。将音频转换为文本后，我们训练了模型来提取查询的意图，找到适当的响应并以自然的人类声音与客户交谈。为了衡量系统的性能，我们使用了召回，精度和F1分数等性能指标。

This paper revisits a fundamental problem in statistical inference from a non-asymptotic theoretical viewpoint $\unicode{x2013}$ the construction of confidence sets. We establish a finite-sample bound for the estimator, characterizing its asymptotic behavior in a non-asymptotic fashion. An important feature of our bound is that its dimension dependency is captured by the effective dimension $\unicode{x2013}$ the trace of the limiting sandwich covariance $\unicode{x2013}$ which can be much smaller than the parameter dimension in some regimes. We then illustrate how the bound can be used to obtain a confidence set whose shape is adapted to the optimization landscape induced by the loss function. Unlike previous works that rely heavily on the strong convexity of the loss function, we only assume the Hessian is lower bounded at optimum and allow it to gradually becomes degenerate. This property is formalized by the notion of generalized self-concordance which originated from convex optimization. Moreover, we demonstrate how the effective dimension can be estimated from data and characterize its estimation accuracy. We apply our results to maximum likelihood estimation with generalized linear models, score matching with exponential families, and hypothesis testing with Rao's score test.

Generative AI has matured to a point where large-scale models can generate text that seems indistinguishable from human-written text and remarkably photorealistic images. Automatically measuring how close the distribution of generated data is to the target real data distribution is a key step in diagnosing existing models and developing better models. We present MAUVE, a family of comparison measures between pairs of distributions such as those encountered in the generative modeling of text or images. These scores are statistical summaries of divergence frontiers capturing two types of errors in generative modeling. We explore four approaches to statistically estimate these scores: vector quantization, non-parametric estimation, classifier-based estimation, and parametric Gaussian approximations. We provide statistical bounds for the vector quantization approach. Empirically, we find that the proposed scores paired with a range of $f$-divergences and statistical estimation methods can quantify the gaps between the distributions of human-written text and those of modern neural language models by correlating with human judgments and identifying known properties of the generated texts. We conclude the paper by demonstrating its applications to other AI domains and discussing practical recommendations.

Event-based vision has been rapidly growing in recent years justified by the unique characteristics it presents such as its high temporal resolutions (~1us), high dynamic range (>120dB), and output latency of only a few microseconds. This work further explores a hybrid, multi-modal, approach for object detection and tracking that leverages state-of-the-art frame-based detectors complemented by hand-crafted event-based methods to improve the overall tracking performance with minimal computational overhead. The methods presented include event-based bounding box (BB) refinement that improves the precision of the resulting BBs, as well as a continuous event-based object detection method, to recover missed detections and generate inter-frame detections that enable a high-temporal-resolution tracking output. The advantages of these methods are quantitatively verified by an ablation study using the higher order tracking accuracy (HOTA) metric. Results show significant performance gains resembled by an improvement in the HOTA from 56.6%, using only frames, to 64.1% and 64.9%, for the event and edge-based mask configurations combined with the two methods proposed, at the baseline framerate of 24Hz. Likewise, incorporating these methods with the same configurations has improved HOTA from 52.5% to 63.1%, and from 51.3% to 60.2% at the high-temporal-resolution tracking rate of 384Hz. Finally, a validation experiment is conducted to analyze the real-world single-object tracking performance using high-speed LiDAR. Empirical evidence shows that our approaches provide significant advantages compared to using frame-based object detectors at the baseline framerate of 24Hz and higher tracking rates of up to 500Hz.

Open-textured terms in written rules are typically settled through interpretive argumentation. Ongoing work has attempted to catalogue the schemes used in such interpretive argumentation. But how can the use of these schemes affect the way in which people actually use and reason over the proper interpretations of open-textured terms? Using the interpretive argument-eliciting game Aporia as our framework, we carried out an empirical study to answer this question. Differing from previous work, we did not allow participants to argue for interpretations arbitrarily, but to only use arguments that fit with a given set of interpretive argument templates. Finally, we analyze the results captured by this new dataset, specifically focusing on practical implications for the development of interpretation-capable artificial reasoners.

Spectral risk objectives - also called $L$-risks - allow for learning systems to interpolate between optimizing average-case performance (as in empirical risk minimization) and worst-case performance on a task. We develop stochastic algorithms to optimize these quantities by characterizing their subdifferential and addressing challenges such as biasedness of subgradient estimates and non-smoothness of the objective. We show theoretically and experimentally that out-of-the-box approaches such as stochastic subgradient and dual averaging are hindered by bias and that our approach outperforms them.

Influence diagnostics such as influence functions and approximate maximum influence perturbations are popular in machine learning and in AI domain applications. Influence diagnostics are powerful statistical tools to identify influential datapoints or subsets of datapoints. We establish finite-sample statistical bounds, as well as computational complexity bounds, for influence functions and approximate maximum influence perturbations using efficient inverse-Hessian-vector product implementations. We illustrate our results with generalized linear models and large attention based models on synthetic and real data.

Adversarial training has been empirically shown to be more prone to overfitting than standard training. The exact underlying reasons still need to be fully understood. In this paper, we identify one cause of overfitting related to current practices of generating adversarial samples from misclassified samples. To address this, we propose an alternative approach that leverages the misclassified samples to mitigate the overfitting problem. We show that our approach achieves better generalization while having comparable robustness to state-of-the-art adversarial training methods on a wide range of computer vision, natural language processing, and tabular tasks.

Adversarial training is widely acknowledged as the most effective defense against adversarial attacks. However, it is also well established that achieving both robustness and generalization in adversarially trained models involves a trade-off. The goal of this work is to provide an in depth comparison of different approaches for adversarial training in language models. Specifically, we study the effect of pre-training data augmentation as well as training time input perturbations vs. embedding space perturbations on the robustness and generalization of BERT-like language models. Our findings suggest that better robustness can be achieved by pre-training data augmentation or by training with input space perturbation. However, training with embedding space perturbation significantly improves generalization. A linguistic correlation analysis of neurons of the learned models reveal that the improved generalization is due to `more specialized' neurons. To the best of our knowledge, this is the first work to carry out a deep qualitative analysis of different methods of generating adversarial examples in adversarial training of language models.

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.