Search for: All records

Low efficiency in recovering low-grade heat remains unresolved despite decades of attempts. In this research, we designed and fabricated a novel thermo-osmotic ionogel (TOI) composite to recover low-grade heat to generate electric power through a thermo-induced ion gradient and selective ion diffusion. The TOI composite was assembled with a crystalline ionogel (polymer-confined LiNO 3 –3H 2 O) film, ion selective membrane, and hydrogel film. With a 90 °C heat supply, the single TOI composite produced a high open-circuit voltage of 0.52 V, a differential thermal voltage of ∼26 mV K −1 , a peak power density of 0.4 W m −2 , and a ground-breaking peak energy conversion efficiency of 11.17%. Eight pieces of such a TOI composite were connected in series, demonstrating an open-circuit voltage of 3.25 volts. Such a TOI system was also demonstrated to harvest body temperature for powering a LED, opening numerous opportunities in powering wearable devices. 

Video-language models (VLMs), large models pre-trained on numerous but noisy video-text pairs from the internet, have revolutionized activity recognition through their remarkable generalization and open-vocabulary capabilities. While complex human activities are often hierarchical and compositional, most existing tasks for evaluating VLMs focus only on high-level video understanding, making it difficult to accurately assess and interpret the ability of VLMs to understand complex and fine-grained human activities. Inspired by the recently proposed MOMA framework, we define activity graphs as a single universal representation of human activities that encompasses video understanding at the activity, sub10 activity, and atomic action level. We redefine activity parsing as the overarching task of activity graph generation, requiring understanding human activities across all three levels. To facilitate the evaluation of models on activity parsing, we introduce MOMA-LRG (Multi-Object Multi-Actor Language-Refined Graphs), a large dataset of complex human activities with activity graph annotations that can be readily transformed into natural language sentences. Lastly, we present a model-agnostic and lightweight approach to adapting and evaluating VLMs by incorporating structured knowledge from activity graphs into VLMs, addressing the individual limitations of language and graphical models. We demonstrate a strong performance on activity parsing and few-shot video classification, and our framework is intended to foster future research in the joint modeling of videos, graphs, and language.