The past years have witnessed a proliferation of large language models (LLMs). Yet, automated and unbiased evaluation of LLMs is challenging due to the inaccuracy of standard metrics in reflecting human preferences and the inefficiency in sampling informative and diverse test examples. While human evaluation remains the gold standard, it is expensive and time-consuming, especially when dealing with a large number of testing samples. To address this problem, we propose a sample-efficient human evaluation method based on MAximum Discrepancy (MAD) competition. MAD automatically selects a small set of informative and diverse instructions, each adapted to two LLMs, whose responses are subject to three-alternative forced choice by human subjects. The pairwise comparison results are then aggregated into a global ranking using the Elo rating system. We select eight representative LLMs and compare them in terms of four skills: knowledge understanding, mathematical reasoning, writing, and coding. Experimental results show that the proposed method achieves a reliable and sensible ranking of LLMs' capabilities, identifies their relative strengths and weaknesses, and offers valuable insights for further LLM advancement.

Crossword puzzles are popular linguistic games often used as tools to engage students in learning. Educational crosswords are characterized by less cryptic and more factual clues that distinguish them from traditional crossword puzzles. Despite there exist several publicly available clue-answer pair databases for traditional crosswords, educational clue-answer pairs datasets are missing. In this article, we propose a methodology to build educational clue generation datasets that can be used to instruct Large Language Models (LLMs). By gathering from Wikipedia pages informative content associated with relevant keywords, we use Large Language Models to automatically generate pedagogical clues related to the given input keyword and its context. With such an approach, we created clue-instruct, a dataset containing 44,075 unique examples with text-keyword pairs associated with three distinct crossword clues. We used clue-instruct to instruct different LLMs to generate educational clues from a given input content and keyword. Both human and automatic evaluations confirmed the quality of the generated clues, thus validating the effectiveness of our approach.

This property is known as mode connectivity. Continual learning poses a fundamental challenge for One feature that may be considered when studying this phenomenon modern machine learning systems, requiring models to is the permutation invariance of neural networks adapt to new tasks while retaining knowledge from previous [12]. Neurons or kernels of network layers can be permuted ones. Addressing this challenge necessitates the development and, if neighboring layers' outputs and inputs are adjusted, of efficient algorithms capable of learning from data one can obtain a solution that has the same properties as streams and accumulating knowledge over time. This paper the original model but lies in a completely different part of proposes a novel approach to continual learning utilizing the loss landscape. Considering this fact, one may conclude the weight consolidation method. Our method, a simple that the abundance of local minima in the loss landscape of yet powerful technique, enhances robustness against neural networks results from permutation invariance. In a catastrophic forgetting by interpolating between old and follow-up work, Ainsworth et al. [1] showed how to find new model weights after each novel task, effectively merging permutations of weights that allow for a linear interpolation two models to facilitate exploration of local minima of weights with low or even near zero barriers.

Frontier large language models (LLMs) are developed by researchers and practitioners with skewed cultural backgrounds and on datasets with skewed sources. However, LLMs' (lack of) multicultural knowledge cannot be effectively assessed with current methods for developing benchmarks. Existing multicultural evaluations primarily rely on expensive and restricted human annotations or potentially outdated internet resources. Thus, they struggle to capture the intricacy, dynamics, and diversity of cultural norms. LLM-generated benchmarks are promising, yet risk propagating the same biases they are meant to measure. To synergize the creativity and expert cultural knowledge of human annotators and the scalability and standardizability of LLM-based automation, we introduce CulturalTeaming, an interactive red-teaming system that leverages human-AI collaboration to build truly challenging evaluation dataset for assessing the multicultural knowledge of LLMs, while improving annotators' capabilities and experiences. Our study reveals that CulturalTeaming's various modes of AI assistance support annotators in creating cultural questions, that modern LLMs fail at, in a gamified manner. Importantly, the increased level of AI assistance (e.g., LLM-generated revision hints) empowers users to create more difficult questions with enhanced perceived creativity of themselves, shedding light on the promises of involving heavier AI assistance in modern evaluation dataset creation procedures. Through a series of 1-hour workshop sessions, we gather CULTURALBENCH-V0.1, a compact yet high-quality evaluation dataset with users' red-teaming attempts, that different families of modern LLMs perform with accuracy ranging from 37.7% to 72.2%, revealing a notable gap in LLMs' multicultural proficiency.

Reading comprehension continues to be a crucial research focus in the NLP community. Recent advances in Machine Reading Comprehension (MRC) have mostly centered on literal comprehension, referring to the surface-level understanding of content. In this work, we focus on the next level - interpretive comprehension, with a particular emphasis on inferring the themes of a narrative text. We introduce the first dataset specifically designed for interpretive comprehension of educational narratives, providing corresponding well-edited theme texts. The dataset spans a variety of genres and cultural origins and includes human-annotated theme keywords with varying levels of granularity. We further formulate NLP tasks under different abstractions of interpretive comprehension toward the main idea of a story. After conducting extensive experiments with state-of-the-art methods, we found the task to be both challenging and significant for NLP research.

Federated learning (FL) enables collaborative model training among multiple clients without raw data exposure. However, recent studies have shown that clients' private training data can be reconstructed from the gradients they share in FL, known as gradient inversion attacks (GIAs). While GIAs have demonstrated effectiveness under \emph{ideal settings and auxiliary assumptions}, their actual efficacy against \emph{practical FL systems} remains under-explored. To address this gap, we conduct a comprehensive study on GIAs in this work. We start with a survey of GIAs that establishes a milestone to trace their evolution and develops a systematization to uncover their inherent threats. Specifically, we categorize the auxiliary assumptions used by existing GIAs based on their practical accessibility to potential adversaries. To facilitate deeper analysis, we highlight the challenges that GIAs face in practical FL systems from three perspectives: \textit{local training}, \textit{model}, and \textit{post-processing}. We then perform extensive theoretical and empirical evaluations of state-of-the-art GIAs across diverse settings, utilizing eight datasets and thirteen models. Our findings indicate that GIAs have inherent limitations when reconstructing data under practical local training settings. Furthermore, their efficacy is sensitive to the trained model, and even simple post-processing measures applied to gradients can be effective defenses. Overall, our work provides crucial insights into the limited effectiveness of GIAs in practical FL systems. By rectifying prior misconceptions, we hope to inspire more accurate and realistic investigations on this topic.

Scaling laws describe the relationship between the size of language models and their capabilities. Unlike prior studies that evaluate a model's capability via loss or benchmarks, we estimate the number of knowledge bits a model stores. We focus on factual knowledge represented as tuples, such as (USA, capital, Washington D.C.) from a Wikipedia page. Through multiple controlled datasets, we establish that language models can and only can store 2 bits of knowledge per parameter, even when quantized to int8, and such knowledge can be flexibly extracted for downstream applications. Consequently, a 7B model can store 14B bits of knowledge, surpassing the English Wikipedia and textbooks combined based on our estimation. More broadly, we present 12 results on how (1) training duration, (2) model architecture, (3) quantization, (4) sparsity constraints such as MoE, and (5) data signal-to-noise ratio affect a model's knowledge storage capacity. Notable insights include: * The GPT-2 architecture, with rotary embedding, matches or even surpasses LLaMA/Mistral architectures in knowledge storage, particularly over shorter training durations. This arises because LLaMA/Mistral uses GatedMLP, which is less stable and harder to train. * Prepending training data with domain names (e.g., wikipedia.org) significantly increases a model's knowledge capacity. Language models can autonomously identify and prioritize domains rich in knowledge, optimizing their storage capacity.

We investigate how to use AI-based analytics to support design education. The analytics at hand measure multiscale design, that is, students' use of space and scale to visually and conceptually organize their design work. With the goal of making the analytics intelligible to instructors, we developed a research artifact integrating a design analytics dashboard with design instances, and the design environment that students use to create them. We theorize about how Suchman's notion of mutual intelligibility requires contextualized investigation of AI in order to develop findings about how analytics work for people. We studied the research artifact in 5 situated course contexts, in 3 departments. A total of 236 students used the multiscale design environment. The 9 instructors who taught those students experienced the analytics via the new research artifact. We derive findings from a qualitative analysis of interviews with instructors regarding their experiences. Instructors reflected on how the analytics and their presentation in the dashboard have the potential to affect design education. We develop research implications addressing: (1) how indexing design analytics in the dashboard to actual design work instances helps design instructors reflect on what they mean and, more broadly, is a technique for how AI-based design analytics can support instructors' assessment and feedback experiences in situated course contexts; and (2) how multiscale design analytics, in particular, have the potential to support design education. By indexing, we mean linking which provides context, here connecting the numbers of the analytics with visually annotated design work instances.

The body structures of tendon-driven musculoskeletal humanoids are complex, and accurate modeling is difficult, because they are made by imitating the body structures of human beings. For this reason, we have not been able to move them accurately like ordinary humanoids driven by actuators in each axis, and large internal muscle tension and slack of tendon wires have emerged by the model error between its geometric model and the actual robot. Therefore, we construct a joint-muscle mapping (JMM) using a neural network (NN), which expresses a nonlinear relationship between joint angles and muscle lengths, and aim to move tendon-driven musculoskeletal humanoids accurately by updating the JMM online from data of the actual robot. In this study, the JMM is updated online by using the vision of the robot so that it moves to the correct position (Vision Updater). Also, we execute another update to modify muscle antagonisms correctly (Antagonism Updater). By using these two updaters, the error between the target and actual joint angles decrease to about 40% in 5 minutes, and we show through a manipulation experiment that the tendon-driven musculoskeletal humanoid Kengoro becomes able to move as intended. This novel system can adapt to the state change and growth of robots, because it updates the JMM online successively.

This paper investigates the influence of anthropomorphized descriptions of so-called "AI" (artificial intelligence) systems on people's self-assessment of trust in the system. Building on prior work, we define four categories of anthropomorphization (1. Properties of a cognizer, 2. Agency, 3. Biological metaphors, and 4. Properties of a communicator). We use a survey-based approach (n=954) to investigate whether participants are likely to trust one of two (fictitious) "AI" systems by randomly assigning people to see either an anthropomorphized or a de-anthropomorphized description of the systems. We find that participants are no more likely to trust anthropomorphized over de-anthropmorphized product descriptions overall. The type of product or system in combination with different anthropomorphic categories appears to exert greater influence on trust than anthropomorphizing language alone, and age is the only demographic factor that significantly correlates with people's preference for anthropomorphized or de-anthropomorphized descriptions. When elaborating on their choices, participants highlight factors such as lesser of two evils, lower or higher stakes contexts, and human favoritism as driving motivations when choosing between product A and B, irrespective of whether they saw an anthropomorphized or a de-anthropomorphized description of the product. Our results suggest that "anthropomorphism" in "AI" descriptions is an aggregate concept that may influence different groups differently, and provide nuance to the discussion of whether anthropomorphization leads to higher trust and over-reliance by the general public in systems sold as "AI".