Motion trajectories offer reliable references for physics-based motion learning but suffer from sparsity, particularly in regions that lack sufficient data coverage. To address this challenge, we introduce a self-supervised, structured representation and generation method that extracts spatial-temporal relationships in periodic or quasi-periodic motions. The motion dynamics in a continuously parameterized latent space enable our method to enhance the interpolation and generalization capabilities of motion learning algorithms. The motion learning controller, informed by the motion parameterization, operates online tracking of a wide range of motions, including targets unseen during training. With a fallback mechanism, the controller dynamically adapts its tracking strategy and automatically resorts to safe action execution when a potentially risky target is proposed. By leveraging the identified spatial-temporal structure, our work opens new possibilities for future advancements in general motion representation and learning algorithms. The availability of reference trajectories, such as motion capture data, has significantly propelled the advancement of motion learning techniques (Peng et al., 2018; Bergamin et al., 2019; Peng et al., 2021; 2022; Starke et al., 2022; Li et al., 2023b;a). However, it is difficult to generalize policies using these techniques to motions outside the distribution of the available data (Peng et al., 2020; Li et al., 2023a). A core reason is that, while the trajectories in the data itself are induced by some dynamics of the system, the learned policies are typically trained to only replicate the data, instead of understanding the underlying dynamics structure. In other words, the policies attempt to memorize the trajectory instances rather than learn to predict them systematically. Moreover, the high nonlinearity and the embedded high-level similarity hinder datadriven methods from effectively identifying and modeling the dynamics of motion patterns (Peng et al., 2018). Therefore, addressing these challenges requires systematic understanding and leveraging the structured nature of the motion space. Instead of handling raw motion trajectories in long-horizon, high-dimensional state space, structured representation methods introduce certain inductive biases during training and offer an efficient approach to managing complex movements (Min & Chai, 2012; Lee et al., 2021). These methods focus on extracting the essential features and temporal dependencies of motions, enabling more effective and compact representations (Lee et al., 2010; Levine et al., 2012). The ability to understand and capture the spatial-temporal structure of the motion space offers enhanced interpolation and generalization capabilities that can augment training datasets and improve the effectiveness of motion generation algorithms (Holden et al., 2017; Iscen et al., 2018; Ibarz et al., 2021).

A successful negotiation demands a deep comprehension of the conversation context, Theory-of-Mind (ToM) skills to infer the partner's motives, as well as strategic reasoning and effective communication, making it challenging for automated systems. Given the remarkable performance of LLMs across a variety of NLP tasks, in this work, we aim to understand how LLMs can advance different aspects of negotiation research, ranging from designing dialogue systems to providing pedagogical feedback and scaling up data collection practices. To this end, we devise a methodology to analyze the multifaceted capabilities of LLMs across diverse dialogue scenarios covering all the time stages of a typical negotiation interaction. Our analysis adds to the increasing evidence for the superiority of GPT-4 across various tasks while also providing insights into specific tasks that remain difficult for LLMs. For instance, the models correlate poorly with human players when making subjective assessments about the negotiation dialogues and often struggle to generate responses that are contextually appropriate as well as strategically advantageous.

Sequence labeling models often benefit from incorporating external knowledge. However, this practice introduces data heterogeneity and complicates the model with additional modules, leading to increased expenses for training a high-performing model. To address this challenge, we propose a two-stage curriculum learning (TCL) framework specifically designed for sequence labeling tasks. The TCL framework enhances training by gradually introducing data instances from easy to hard, aiming to improve both performance and training speed. Furthermore, we explore different metrics for assessing the difficulty levels of sequence labeling tasks. Through extensive experimentation on six Chinese word segmentation (CWS) and Part-of-speech tagging (POS) datasets, we demonstrate the effectiveness of our model in enhancing the performance of sequence labeling models. Additionally, our analysis indicates that TCL accelerates training and alleviates the slow training problem associated with complex models.

Jay fell in love with math at boarding school after a supportive physics teacher introduced him to the joy of complex calculus. He went on to study physics and math in college, hoping to one day similarly pass on what he'd learned to a new generation. That chance came in October 2022, when 25-year-old Jay answered a job listing seeking a math expert to grade equations through an online platform. But he would not be inspiring budding young mathematicians like his past self. He would instead be training an artificial intelligence system that may eventually make his expertise obsolete.

With the rapid development of big data and AI technology, programming is in high demand and has become an essential skill for students. Meanwhile, researchers also focus on boosting the online judging system's guidance ability to reduce students' dropout rates. Previous studies mainly targeted at enhancing learner engagement on online platforms by providing personalized recommendations. However, two significant challenges still need to be addressed in programming: C1) how to recognize complex programming behaviors; C2) how to capture intrinsic learning patterns that align with the actual learning process. To fill these gaps, in this paper, we propose a novel model called Programming Exercise Recommender with Learning Style (PERS), which simulates learners' intricate programming behaviors. Specifically, since programming is an iterative and trial-and-error process, we first introduce a positional encoding and a differentiating module to capture the changes of consecutive code submissions (which addresses C1). To better profile programming behaviors, we extend the Felder-Silverman learning style model, a classical pedagogical theory, to perceive intrinsic programming patterns. Based on this, we align three latent vectors to record and update programming ability, processing style, and understanding style, respectively (which addresses C2). We perform extensive experiments on two real-world datasets to verify the rationality of modeling programming learning styles and the effectiveness of PERS for personalized programming guidance.

Modern language models rely on the transformer architecture and attention mechanism to perform language understanding and text generation. In this work, we study learning a 1-layer self-attention model from a set of prompts and associated output data sampled from the model. We first establish a precise mapping between the self-attention mechanism and Markov models: Inputting a prompt to the model samples the output token according to a context-conditioned Markov chain (CCMC) which weights the transition matrix of a base Markov chain. Additionally, incorporating positional encoding results in position-dependent scaling of the transition probabilities. Building on this formalism, we develop identifiability/coverage conditions for the prompt distribution that guarantee consistent estimation and establish sample complexity guarantees under IID samples. Finally, we study the problem of learning from a single output trajectory generated from an initial prompt. We characterize an intriguing winner-takes-all phenomenon where the generative process implemented by self-attention collapses into sampling a limited subset of tokens due to its non-mixing nature. This provides a mathematical explanation to the tendency of modern LLMs to generate repetitive text. In summary, the equivalence to CCMC provides a simple but powerful framework to study self-attention and its properties.

We address the growing apprehension that GNNs, in the absence of fairness constraints, might produce biased decisions that disproportionately affect underprivileged groups or individuals. Departing from previous work, we introduce for the first time a method for incorporating the Gini coefficient as a measure of fairness to be used within the GNN framework. Our proposal, GRAPHGINI, works with the two different goals of individual and group fairness in a single system, while maintaining high prediction accuracy. GRAPHGINI enforces individual fairness through learnable attention scores that help in aggregating more information through similar nodes. A heuristic-based maximum Nash social welfare constraint ensures the maximum possible group fairness. Both the individual fairness constraint and the group fairness constraint are stated in terms of a differentiable approximation of the Gini coefficient. This approximation is a contribution that is likely to be of interest even beyond the scope of the problem studied in this paper. Unlike other state-of-the-art, GRAPHGINI automatically balances all three optimization objectives (utility, individual, and group fairness) of the GNN and is free from any manual tuning of weight parameters. Extensive experimentation on real-world datasets showcases the efficacy of GRAPHGINI in making significant improvements in individual fairness compared to all currently available state-of-the-art methods while maintaining utility and group equality.

Modern large language models (LLMs) should generally benefit individuals from various cultural backgrounds around the world. However, most recent advanced generative evaluation benchmarks tailed for LLMs mainly focus on English. To this end, we introduce OMGEval, the first Open-source Multilingual Generative test set that can assess the capability of LLMs in different languages. For each language, OMGEval provides 804 open-ended questions, covering a wide range of important capabilities of LLMs, such as general knowledge, logical reasoning, and so on. Each question is rigorously verified by human annotators. Notably, to sufficiently reflect the compatibility of LLMs in different cultural backgrounds, we perform localization for each non-English language. Specifically, the current version of OMGEval includes 5 languages (i.e., Zh, Ru, Fr, Es, Ar). Following AlpacaEval, we employ GPT-4 as the adjudicator to automatically score different model outputs, which is shown closely related to human evaluation. We evaluate several representative multilingual LLMs on the proposed OMGEval, which we believe will provide a valuable reference for the community to further understand and improve the multilingual capability of LLMs. OMGEval is available at https://github.com/blcuicall/OMGEval.

Deep neural networks (DNNs) have proven to be effective models for accurate Memory Access Prediction (MAP), a critical task in mitigating memory latency through data prefetching. However, existing DNN-based MAP models suffer from the challenges such as significant physical storage space and poor inference latency, primarily due to their large number of parameters. These limitations render them impractical for deployment in real-world scenarios. In this paper, we propose PaCKD, a Pattern-Clustered Knowledge Distillation approach to compress MAP models while maintaining the prediction performance. The PaCKD approach encompasses three steps: clustering memory access sequences into distinct partitions involving similar patterns, training large pattern-specific teacher models for memory access prediction for each partition, and training a single lightweight student model by distilling the knowledge from the trained pattern-specific teachers. We evaluate our approach on LSTM, MLP-Mixer, and ResNet models, as they exhibit diverse structures and are widely used for image classification tasks in order to test their effectiveness in four widely used graph applications. Compared to the teacher models with 5.406M parameters and an F1-score of 0.4626, our student models achieve a 552$\times$ model size compression while maintaining an F1-score of 0.4538 (with a 1.92% performance drop). Our approach yields an 8.70% higher result compared to student models trained with standard knowledge distillation and an 8.88% higher result compared to student models trained without any form of knowledge distillation.

In this paper, we introduce BiMediX, the first bilingual medical mixture of experts LLM designed for seamless interaction in both English and Arabic. Our model facilitates a wide range of medical interactions in English and Arabic, including multi-turn chats to inquire about additional details such as patient symptoms and medical history, multiple-choice question answering, and open-ended question answering. We propose a semi-automated English-to-Arabic translation pipeline with human refinement to ensure high-quality translations. We also introduce a comprehensive evaluation benchmark for Arabic medical LLMs. Furthermore, we introduce BiMed1.3M, an extensive Arabic-English bilingual instruction set covering 1.3 Million diverse medical interactions, resulting in over 632 million healthcare specialized tokens for instruction tuning. Our BiMed1.3M dataset includes 250k synthesized multi-turn doctor-patient chats and maintains a 1:2 Arabic-to-English ratio. Our model outperforms state-of-the-art Med42 and Meditron by average absolute gains of 2.5% and 4.1%, respectively, computed across multiple medical evaluation benchmarks in English, while operating at 8-times faster inference. Moreover, our BiMediX outperforms the generic Arabic-English bilingual LLM, Jais-30B, by average absolute gains of 10% on our Arabic medical benchmark and 15% on bilingual evaluations across multiple datasets. Our project page with source code and trained model is available at https://github.com/mbzuai-oryx/BiMediX .