We investigate complex video question answering via chain-of-evidence reasoning -- identifying sequences of temporal spans from multiple relevant parts of the video, together with visual evidence within them. Existing models struggle with multi-step reasoning as they uniformly sample a fixed number of frames, which can miss critical evidence distributed nonuniformly throughout the video. Moreover, they lack the ability to temporally localize such evidence in the broader context of the full video, which is required for answering complex questions. We propose a framework to enhance existing VideoQA datasets with evidence reasoning chains, automatically constructed by searching for optimal intervals of interest in the video with supporting evidence, that maximizes the likelihood of answering a given question. We train our model (VITED) to generate these evidence chains directly, enabling it to both localize evidence windows as well as perform multi-step reasoning across them in long-form video content. We show the value of our evidence-distilled models on a suite of long video QA benchmarks where we outperform state-of-the-art approaches that lack evidence reasoning capabilities.

Foundation models, including language models, e.g., GPT, and vision models, e.g., CLIP, have significantly advanced numerous biomedical tasks. Despite these advancements, the high inference latency and the "overthinking" issues in model inference impair the efficiency and effectiveness of foundation models, thus limiting their application in real-time clinical settings. To address these challenges, we proposed EPEE (Entropy- and Patience-based Early Exiting), a novel hybrid strategy designed to improve the inference efficiency of foundation models. The core idea was to leverage the strengths of entropy-based and patience-based early exiting methods to overcome their respective weaknesses. To evaluate EPEE, we conducted experiments on three core biomedical tasks-classification, relation extraction, and event extraction-using four foundation models (BERT, ALBERT, GPT-2, and ViT) across twelve datasets, including clinical notes and medical images. The results showed that EPEE significantly reduced inference time while maintaining or improving accuracy, demonstrating its adaptability to diverse datasets and tasks. EPEE addressed critical barriers to deploying foundation models in healthcare by balancing efficiency and effectiveness. It potentially provided a practical solution for real-time clinical decision-making with foundation models, supporting reliable and efficient workflows.

In this work, a large area, low cost and flexible polymer/paper-based double touch mode capacitive pressure sensor is demonstrated. Garage fabrication processes are used which only require cutting, taping and assembly of aluminum (Al) coated polyimide (PI) foil, PI tape and double-sided scotch tape. The presented pressure sensor operates in different pressure regions i.e. normal (0 to 7.5 kPa), transition (7.5 to 14.24 kPa), linear (14.24 to 54.9 kPa) and saturation (above 54.9 kPa). The advantages of the demonstrated double touch mode capacitive pressure sensors are low temperature drift, long linear range, high pressure sensitivity, precise pressure measurement and large die area. The linear output along with a high sensitivity range (14.24 to 54.9 kPa pressure range) of the sensor are utilized to wirelessly control the movement of a robotic arm with precise rotation and tilt movement capabilities.

Justin Gerrard speaks quickly, Brian Gerrard speaks slowly. If you met them separately, you would never guess they were brothers. But their oil-and-water partnership helped them create Bae, a dating app for black people. Bae works pretty much like Tinder, but tailor-made for black users. The Gerrards came up with the idea after they realized how difficult it is for black singles to find dates on existing platforms.