This paper reports on a set of three recent experiments utilizing large-scale speech models to evaluate the oral reading fluency (ORF) of students in Ghana. While ORF is a well-established measure of foundational literacy, assessing it typically requires one-on-one sessions between a student and a trained evaluator, a process that is time-consuming and costly. Automating the evaluation of ORF could support better literacy instruction, particularly in education contexts where formative assessment is uncommon due to large class sizes and limited resources. To our knowledge, this research is among the first to examine the use of the most recent versions of large-scale speech models (Whisper V2 wav2vec2.0) for ORF assessment in the Global South. We find that Whisper V2 produces transcriptions of Ghanaian students reading aloud with a Word Error Rate of 13.5. This is close to the model's average WER on adult speech (12.8) and would have been considered state-of-the-art for children's speech transcription only a few years ago. We also find that when these transcriptions are used to produce fully automated ORF scores, they closely align with scores generated by expert human graders, with a correlation coefficient of 0.96. Importantly, these results were achieved on a representative dataset (i.e., students with regional accents, recordings taken in actual classrooms), using a free and publicly available speech model out of the box (i.e., no fine-tuning). This suggests that using large-scale speech models to assess ORF may be feasible to implement and scale in lower-resource, linguistically diverse educational contexts.

With LLMs shifting their role from statistical modeling of language to serving as general-purpose AI agents, how should LLM evaluations change? Arguably, a key ability of an AI agent is to flexibly combine, as needed, the basic skills it has learned. The capability to combine skills plays an important role in (human) pedagogy and also in a paper on emergence phenomena (Arora & Goyal, 2023). This work introduces Skill-Mix, a new evaluation to measure ability to combine skills. Using a list of $N$ skills the evaluator repeatedly picks random subsets of $k$ skills and asks the LLM to produce text combining that subset of skills. Since the number of subsets grows like $N^k$, for even modest $k$ this evaluation will, with high probability, require the LLM to produce text significantly different from any text in the training set. The paper develops a methodology for (a) designing and administering such an evaluation, and (b) automatic grading (plus spot-checking by humans) of the results using GPT-4 as well as the open LLaMA-2 70B model. Administering a version of to popular chatbots gave results that, while generally in line with prior expectations, contained surprises. Sizeable differences exist among model capabilities that are not captured by their ranking on popular LLM leaderboards ("cramming for the leaderboard"). Furthermore, simple probability calculations indicate that GPT-4's reasonable performance on $k=5$ is suggestive of going beyond "stochastic parrot" behavior (Bender et al., 2021), i.e., it combines skills in ways that it had not seen during training. We sketch how the methodology can lead to a Skill-Mix based eco-system of open evaluations for AI capabilities of future models.

Large language models (LLMs) have been widely used as agents to complete different tasks, such as personal assistance or event planning. While most work has focused on cooperation and collaboration between agents, little work explores competition, another important mechanism that fosters the development of society and economy. In this paper, we seek to examine the competition behaviors in LLM-based agents. We first propose a general framework to study the competition between agents. Then, we implement a practical competitive environment using GPT-4 to simulate a virtual town with two types of agents, including restaurant agents and customer agents. Specifically, restaurant agents compete with each other to attract more customers, where the competition fosters them to transform, such as cultivating new operating strategies. The results of our experiments reveal several interesting findings ranging from social learning to Matthew Effect, which aligns well with existing sociological and economic theories. We believe that competition between agents deserves further investigation to help us understand society better. The code will be released soon.

Automatic sign language recognition (SLR) is an important topic within the areas of human-computer interaction and machine learning. On the one hand, it poses a complex challenge that requires the intervention of various knowledge areas, such as video processing, image processing, intelligent systems and linguistics. On the other hand, robust recognition of sign language could assist in the translation process and the integration of hearing-impaired people, as well as the teaching of sign language for the hearing population. SLR systems usually employ Hidden Markov Models, Dynamic Time Warping or similar models to recognize signs. Such techniques exploit the sequential ordering of frames to reduce the number of hypothesis. This paper presents a general probabilistic model for sign classification that combines sub-classifiers based on different types of features such as position, movement and handshape. The model employs a bag-of-words approach in all classification steps, to explore the hypothesis that ordering is not essential for recognition. The proposed model achieved an accuracy rate of 97% on an Argentinian Sign Language dataset containing 64 classes of signs and 3200 samples, providing some evidence that indeed recognition without ordering is possible.

Automatic sign language recognition is an important topic within the areas of human-computer interaction and machine learning. On the one hand, it poses a complex challenge that requires the intervention of various knowledge areas, such as video processing, image processing, intelligent systems and linguistics. On the other hand, robust recognition of sign language could assist in the translation process and the integration of hearing-impaired people. This paper offers two main contributions: first, the creation of a database of handshapes for the Argentinian Sign Language (LSA), which is a topic that has barely been discussed so far. Secondly, a technique for image processing, descriptor extraction and subsequent handshape classification using a supervised adaptation of self-organizing maps that is called ProbSom. This technique is compared to others in the state of the art, such as Support Vector Machines (SVM), Random Forests, and Neural Networks. The database that was built contains 800 images with 16 LSA handshapes, and is a first step towards building a comprehensive database of Argentinian signs. The ProbSom-based neural classifier, using the proposed descriptor, achieved an accuracy rate above 90%.

We study multitask online learning in a setting where agents can only exchange information with their neighbors on an arbitrary communication network. We introduce $\texttt{MT-CO}_2\texttt{OL}$, a decentralized algorithm for this setting whose regret depends on the interplay between the task similarities and the network structure. Our analysis shows that the regret of $\texttt{MT-CO}_2\texttt{OL}$ is never worse (up to constants) than the bound obtained when agents do not share information. On the other hand, our bounds significantly improve when neighboring agents operate on similar tasks. In addition, we prove that our algorithm can be made differentially private with a negligible impact on the regret when the losses are linear. Finally, we provide experimental support for our theory.

Spatio-temporal graph learning is a fundamental problem in the Web of Things era, which enables a plethora of Web applications such as smart cities, human mobility and climate analysis. Existing approaches tackle different learning tasks independently, tailoring their models to unique task characteristics. These methods, however, fall short of modeling intrinsic uncertainties in the spatio-temporal data. Meanwhile, their specialized designs limit their universality as general spatio-temporal learning solutions. In this paper, we propose to model the learning tasks in a unified perspective, viewing them as predictions based on conditional information with shared spatio-temporal patterns. Based on this proposal, we introduce Unified Spatio-Temporal Diffusion Models (USTD) to address the tasks uniformly within the uncertainty-aware diffusion framework. USTD is holistically designed, comprising a shared spatio-temporal encoder and attention-based denoising networks that are task-specific. The shared encoder, optimized by a pre-training strategy, effectively captures conditional spatio-temporal patterns. The denoising networks, utilizing both cross- and self-attention, integrate conditional dependencies and generate predictions. Opting for forecasting and kriging as downstream tasks, we design Gated Attention (SGA) and Temporal Gated Attention (TGA) for each task, with different emphases on the spatial and temporal dimensions, respectively. By combining the advantages of deterministic encoders and probabilistic diffusion models, USTD achieves state-of-the-art performances compared to deterministic and probabilistic baselines in both tasks, while also providing valuable uncertainty estimates.

Embodied conversational agents (ECAs) are paradigms of conversational user interfaces in the form of embodied characters. While ECAs offer various manipulable features, this paper focuses on a study conducted to explore two distinct levels of presentation realism. The two agent versions are photorealistic and animated. The study aims to provide insights and design suggestions for speech-enabled ECAs within serious game environments. A within-subjects, two-by-two factorial design was employed for this research with a cohort of 36 participants balanced for gender. The results showed that both the photorealistic and the animated versions were perceived as highly usable, with overall mean scores of 5.76 and 5.71, respectively. However, 69.4 per cent of the participants stated they preferred the photorealistic version, 25 per cent stated they preferred the animated version and 5.6 per cent had no stated preference. The photorealistic agents were perceived as more realistic and human-like, while the animated characters made the task feel more like a game. Even though the agents' realism had no significant effect on usability, it positively influenced participants' perceptions of the agent. This research aims to lay the groundwork for future studies on ECA realism's impact in serious games across diverse contexts.

Current tools for machine learning fairness only admit a limited range of fairness definitions and have seen little integration with automatic differentiation libraries, despite the central role these libraries play in modern machine learning pipelines. We introduce a framework of fairness regularization terms (fairrets) which quantify bias as modular objectives that are easily integrated in automatic differentiation pipelines. By employing a general definition of fairness in terms of linear-fractional statistics, a wide class of fairrets can be computed efficiently. Experiments show the behavior of their gradients and their utility in enforcing fairness with minimal loss of predictive power compared to baselines. Our contribution includes a PyTorch implementation of the fairret framework.

Despite the success of transformers on various computer vision tasks, they suffer from excessive memory and computational cost. Some works present dynamic vision transformers to accelerate inference by pruning redundant tokens. A key to improving token pruning is using well-trained models as initialization for faster convergence and better performance. However, current base models usually adopt full image training, i.e., using full images as inputs and keeping the whole feature maps through the forward process, which causes inconsistencies with dynamic models that gradually reduce tokens, including calculation pattern, information amount and token selection strategy inconsistencies. Inspired by MAE which performs masking and reconstruction self-supervised task, we devise masked fine-tuning to bridge the gaps between pre-trained base models used for initialization and token pruning based dynamic vision transformers, by masking image patches and predicting the image class label based on left unmasked patches. Extensive experiments on ImageNet demonstrate that base models via masked fine-tuning gain strong occlusion robustness and ability against information loss. With this better initialization, Dynamic ViT achieves higher accuracies, especially under large token pruning ratios (e.g., 81.9% vs. 81.3%, and 62.3% vs. 58.9% for DeiT based Dynamic ViT/0.8 and Dynamic ViT/0.3). Moreover, we apply our method into different token pruning based dynamic vision transformers, different pre-trained models and randomly initialized models to demonstrate the generalization ability.