Understanding sports is crucial for the advancement of Natural Language Processing (NLP) due to its intricate and dynamic nature. Reasoning over complex sports scenarios has posed significant challenges to current NLP technologies which require advanced cognitive capabilities. Toward addressing the limitations of existing benchmarks on sports understanding in the NLP field, we extensively evaluated mainstream large language models for various sports tasks. Our evaluation spans from simple queries on basic rules and historical facts to complex, context-specific reasoning, leveraging strategies from zero-shot to few-shot learning, and chain-of-thought techniques. In addition to unimodal analysis, we further assessed the sports reasoning capabilities of mainstream video language models to bridge the gap in multimodal sports understanding benchmarking. Our findings highlighted the critical challenges of sports understanding for NLP. We proposed a new benchmark based on a comprehensive overview of existing sports datasets and provided extensive error analysis which we hope can help identify future research priorities in this field.

Understanding the extent to which Chain-of-Thought (CoT) generations align with a large language model's (LLM) internal computations is critical for deciding whether to trust an LLM's output. As a proxy for CoT faithfulness, Lanham et al. (2023) propose a metric that measures a model's dependence on its CoT for producing an answer. Within a single family of proprietary models, they find that LLMs exhibit a scaling-then-inverse-scaling relationship between model size and their measure of faithfulness, and that a 13 billion parameter model exhibits increased faithfulness compared to models ranging from 810 million to 175 billion parameters in size. We evaluate whether these results generalize as a property of all LLMs. We replicate the experimental setup in their section focused on scaling experiments with three different families of models and, under specific conditions, successfully reproduce the scaling trends for CoT faithfulness they report. However, after normalizing the metric to account for a model's bias toward certain answer choices, unfaithfulness drops significantly for smaller less-capable models. This normalized faithfulness metric is also strongly correlated ($R^2$=0.74) with accuracy, raising doubts about its validity for evaluating faithfulness.

This work investigates the in-context learning abilities of pretrained large language models (LLMs) when instructed to translate text from a low-resource language into a high-resource language as part of an automated machine translation pipeline. We conduct a set of experiments translating Southern Quechua to Spanish and examine the informativity of various types of information retrieved from a constrained database of digitized pedagogical materials (dictionaries and grammar lessons) and parallel corpora. Using both automatic and human evaluation of model output, we conduct ablation studies that manipulate (1) context type (morpheme translations, grammar descriptions, and corpus examples), (2) retrieval methods (automated vs. manual), and (3) model type. Our results suggest that even relatively small LLMs are capable of utilizing prompt context for zero-shot low-resource translation when provided a minimally sufficient amount of relevant linguistic information. However, the variable effects of prompt type, retrieval method, model type, and language-specific factors highlight the limitations of using even the best LLMs as translation systems for the majority of the world's 7,000+ languages and their speakers.

The use of Retrieval-Augmented Generation (RAG) has improved Large Language Models (LLMs) in collaborating with external data, yet significant challenges exist in real-world scenarios. In areas such as academic literature and finance question answering, data are often found in raw text and tables in HTML or PDF formats, which can be lengthy and highly unstructured. In this paper, we introduce a benchmark suite, namely Unstructured Document Analysis (UDA), that involves 2,965 real-world documents and 29,590 expert-annotated Q&A pairs. We revisit popular LLM- and RAG-based solutions for document analysis and evaluate the design choices and answer qualities across multiple document domains and diverse query types. Our evaluation yields interesting findings and highlights the importance of data parsing and retrieval. We hope our benchmark can shed light and better serve real-world document analysis applications. The benchmark suite and code can be found at https://github.com/qinchuanhui/UDA-Benchmark.

Accurately modeling idiomatic or non-compositional language has been a longstanding challenge in Natural Language Processing (NLP). This is partly because these expressions do not derive their meanings solely from their constituent words, but also due to the scarcity of relevant data resources, and their impact on the performance of downstream tasks such as machine translation and simplification. In this paper we propose an approach to model idiomaticity effectively using a triplet loss that incorporates the asymmetric contribution of components words to an idiomatic meaning for training language models by using adaptive contrastive learning and resampling miners to build an idiomatic-aware learning objective. Our proposed method is evaluated on a SemEval challenge and outperforms previous alternatives significantly in many metrics.

With the increasing use of black-box Machine Learning (ML) techniques in critical applications, there is a growing demand for methods that can provide transparency and accountability for model predictions. As a result, a large number of local explainability methods for black-box models have been developed and popularized. However, machine learning explanations are still hard to evaluate and compare due to the high dimensionality, heterogeneous representations, varying scales, and stochastic nature of some of these methods. Topological Data Analysis (TDA) can be an effective method in this domain since it can be used to transform attributions into uniform graph representations, providing a common ground for comparison across different explanation methods. We present a novel topology-driven visual analytics tool, Mountaineer, that allows ML practitioners to interactively analyze and compare these representations by linking the topological graphs back to the original data distribution, model predictions, and feature attributions. Mountaineer facilitates rapid and iterative exploration of ML explanations, enabling experts to gain deeper insights into the explanation techniques, understand the underlying data distributions, and thus reach well-founded conclusions about model behavior. Furthermore, we demonstrate the utility of Mountaineer through two case studies using real-world data. In the first, we show how Mountaineer enabled us to compare black-box ML explanations and discern regions of and causes of disagreements between different explanations. In the second, we demonstrate how the tool can be used to compare and understand ML models themselves. Finally, we conducted interviews with three industry experts to help us evaluate our work.

Countless domains rely on Machine Learning (ML) models, including safety-critical domains, such as autonomous driving, which this paper focuses on. While the black box nature of ML is simply a nuisance in some domains, in safety-critical domains, this makes ML models difficult to trust. To fully utilize ML models in safety-critical domains, it would be beneficial to have a method to improve trust in model robustness and accuracy without human experts checking each decision. This research proposes a method to increase trust in ML models used in safety-critical domains by ensuring the robustness and completeness of the model's training dataset. Because ML models embody what they are trained with, ensuring the completeness of training datasets can help to increase the trust in the training of ML models. To this end, this paper proposes the use of a domain ontology and an image quality characteristic ontology to validate the domain completeness and image quality robustness of a training dataset. This research also presents an experiment as a proof of concept for this method, where ontologies are built for the emergency road vehicle domain.

Any speech recognition system must learn to recognize the intended words regardless of the various ways in which those words may be pronounced. A substantial amount of the variability in speech is systematic, arising from phonological processes occurring in predictable environments. One such process is place assimilation, where phonemes adopt the articulation place of adjacent phonemes. For instance, the word pair clean pan is frequently pronounced as clea[m] pan, with the wordfinal coronal /n/ in clean assimilating to the subsequent labial [p] in pan. This is a simple yet common phonological process across the world's languages (Hura, Lindblom, and Diehl 1992). In English, it occurs for coronal segments (e.g., /t/, /d/, /n/) that are followed by noncoronals, such as labials (e.g., [p], [b], [m]) or velars (e.g., [k], [g], [N]). Human listeners are able to infer the underlying /n/ when exposed to assimilated inputs like clea[m] pan, allowing them to perceive the intended word clean. This phenomenon is referred to as compensation for assimilation and happens automatically-- that is, humans compensate without conscious awareness of the assimilation itself. Psycholinguistic research has used controlled stimuli to investigate the mechanism behind this process.

Sparse attention can effectively mitigate the significant memory and throughput demands of Large Language Models (LLMs) in long contexts. Existing methods typically employ a uniform sparse attention mask, applying the same sparse pattern across different attention heads and input lengths. However, this uniform approach fails to capture the diverse attention patterns inherent in LLMs, ignoring their distinct accuracy-latency trade-offs. To address this challenge, we propose the Mixture of Attention (MoA), which automatically tailors distinct sparse attention configurations to different heads and layers. MoA constructs and navigates a search space of various attention patterns and their scaling rules relative to input sequence lengths. It profiles the model, evaluates potential configurations, and pinpoints the optimal sparse attention compression plan. MoA adapts to varying input sizes, revealing that some attention heads expand their focus to accommodate longer sequences, while other heads consistently concentrate on fixed-length local contexts. Experiments show that MoA increases the effective context length by $3.9\times$ with the same average attention span, boosting retrieval accuracy by $1.5-7.1\times$ over the uniform-attention baseline across Vicuna-7B, Vicuna-13B, and Llama3-8B models. Moreover, MoA narrows the capability gaps between sparse and dense models, reducing the maximum relative performance drop from $9\%-36\%$ to within $5\%$ across two long-context understanding benchmarks. MoA achieves a $1.2-1.4\times$ GPU memory reduction and boosts decode throughput by $5.5-6.7 \times$ for 7B and 13B dense models on a single GPU, with minimal impact on performance.

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental condition characterized by inattention, hyperactivity, and impulsivity, which can significantly impact an individual's daily functioning and quality of life. Occupational therapy plays a crucial role in managing ADHD by fostering the development of skills needed for daily living and enhancing an individual's ability to participate fully in school, home, and social situations. Recent studies highlight the potential of integrating Large Language Models (LLMs) like ChatGPT and Socially Assistive Robots (SAR) to improve psychological treatments. This integration aims to overcome existing limitations in mental health therapy by providing tailored support and adapting to the unique needs of this sensitive group. However, there remains a significant gap in research exploring the combined use of these advanced technologies in ADHD therapy, suggesting an opportunity for novel therapeutic approaches. Thus, we integrated two advanced language models, ChatGPT-4 Turbo and Claude-3 Opus, into a robotic assistant to explore how well each model performs in robot-assisted interactions. Additionally, we have compared their performance in a simulated therapy scenario to gauge their effectiveness against a clinically validated customized model. The results of this study show that ChatGPT-4 Turbo excelled in performance and responsiveness, making it suitable for time-sensitive applications. Claude-3 Opus, on the other hand, showed strengths in understanding, coherence, and ethical considerations, prioritizing safe and engaging interactions. Both models demonstrated innovation and adaptability, but ChatGPT-4 Turbo offered greater ease of integration and broader language support. The selection between them hinges on the specific demands of ADHD therapy.