Structure reasoning is a fundamental capability of large language models (LLMs), enabling them to reason about structured commonsense and answer multi-hop questions. However, existing benchmarks for structure reasoning mainly focus on horizontal and coordinate structures (e.g. graphs), overlooking the hierarchical relationships within them. Hierarchical structure reasoning is crucial for human cognition, particularly in memory organization and problem-solving. It also plays a key role in various real-world tasks, such as information extraction and decision-making. To address this gap, we propose HiBench, the first framework designed to systematically benchmark the hierarchical reasoning capabilities of LLMs from initial structure generation to final proficiency assessment. It encompasses six representative scenarios, covering both fundamental and practical aspects, and consists of 30 tasks with varying hierarchical complexity, totaling 39,519 queries. To evaluate LLMs comprehensively, we develop five capability dimensions that depict different facets of hierarchical structure understanding. Through extensive evaluation of 20 LLMs from 10 model families, we reveal key insights into their capabilities and limitations: 1) existing LLMs show proficiency in basic hierarchical reasoning tasks; 2) they still struggle with more complex structures and implicit hierarchical representations, especially in structural modification and textual reasoning. Based on these findings, we create a small yet well-designed instruction dataset, which enhances LLMs' performance on HiBench by an average of 88.84% (Llama-3.1-8B) and 31.38% (Qwen2.5-7B) across all tasks. The HiBench dataset and toolkit are available at https://github.com/jzzzzh/HiBench to encourage evaluation.

Automatic detection of backchannel has great potential to enhance artificial mediators, which indicate listeners' attention and agreement in human communication. It is often expressed by subtle non-verbal cues that occur briefly and sparsely. Focusing on identifying and locating these subtle cues (i.e., their occurrence moment and the involved body parts), this paper proposes a novel approach for backchannel detection. In particular, our model utilizes temporal- and modality-attention modules to determine and lead the model to pay more attention to both the indicative moment and the accompanying body parts at that specific time. It achieves an accuracy of 68.6% on the testing set in MultiMediate'23 backchannel detection challenge, outperforming the counterparts. Furthermore, we conducted an ablation study to thoroughly understand the contributions of our model. This study underscores the effectiveness of our selection of modality inputs and the importance of the two attention modules in our model.

Real-time engagement estimation holds significant potential across various research areas, particularly in the realm of human-computer interaction. It empowers artificial agents to dynamically adjust their responses based on user engagement levels, fostering more intuitive and immersive interactions. Despite the strides in automating real-time engagement estimation, the task remains challenging in real-world settings, especially when handling multi-modal human social signals. Capitalizing on human body and audio signals, this paper explores the appropriate feature representations of different modalities and effective modelling of dual conversations. This results in a novel and efficient multi-modal engagement detection model.We thoroughly evaluated our method in the MultiMediate'23 grand challenge. It performs consistently, with a notable improvement over the baseline model. Specifically, while the baseline achieves a concordance correlation coefficient (CCC) of 0.59, our approach yields a CCC of 0.70, suggesting its promising efficacy in real-life engagement detection.

Have you ever picked up a stuffed toy and pretended to play with it in your childhood? We are motivated by the novel use of stuffed toys in enhancing extended reality interaction. A key goal of extended reality is to induce the feeling of presence in its users. Naturally mapped control interface has been shown to enhance presence. The literature also indicates that a high degree of freedom tracking is important to extended reality. Based on these observations, we show that a free-form naturally mapped control interface is well-motivated via a theoretical contextualization. We explore the possibility of building such a controller in the form of stuffed toys. To realize stuffed toys as controllers, a novel soft-pose estimator empowered by cage-based deformation is proposed. It is shown to be effective in tracking the poses and deformations of real soft objects even by training with synthetic data only. Three gameplay prototypes are developed to demonstrate that interactive play can be enabled by the soft-pose estimator. They also form the basis for two user studies that validate the success of tracking stuffed toys with the soft-pose estimator for interactive play.