Search for: All records

Diffusion models (DMs) are a class of generative models that allow sampling from a distribution learned over a training set. When applied to solving inverse problems, the reverse sampling steps are modified to approximately sample from a measurement-conditioned distribution. However, these modifications may be unsuitable for certain settings (e.g., presence of measurement noise) and non-linear tasks, as they often struggle to correct errors from earlier steps and generally require a large number of optimization and/or sampling steps. To address these challenges, we state three conditions for achieving measurement-consistent diffusion trajectories. Building on these conditions, we propose a new optimization-based sampling method that not only enforces standard data manifold measurement consistency and forward diffusion consistency, as seen in previous studies, but also incorporates our proposed step-wise and network-regularized backward diffusion consistency that maintains a diffusion trajectory by optimizing over the input of the pre-trained model at every sampling step. By enforcing these conditions (implicitly or explicitly), our sampler requires significantly fewer reverse steps. Therefore, we refer to our method as Step-wise Triple- Consistent Sampling (SITCOM). Compared to SOTA baselines, our experiments across several linear and non-linear tasks (with natural and medical images) demonstrate that SITCOM achieves competitive or superior results in terms of standard similarity metrics and run-time. 

There is a need to develop new and sustainable agricultural technologies to help provide global food security, and nanoscale materials show promising results in this area. In this study, mesoporous silica nanoparticles (MSNs) and chitosan-coated mesoporous silica nanoparticles (CTS-MSNs) were synthesized and applied to soybeans (Glycine max) by two different strategies in greenhouse and field studies to study the role of dissolved silicic acid and chitosan in enhancing plant growth and suppressing disease damage caused by Fusarium virguliforme. Plant growth and health were assessed by measuring the soybean biomass and chlorophyll content in both healthy and Fusarium-infected plants at harvest. In the greenhouse study, foliar and seed applications with 250 mg/L nanoparticle treatments were compared. A single seed treatment of MSNs reduced disease severity by 30% and increased chlorophyll content in both healthy and infected plants by 12%. Based on greenhouse results, seed application was used in the follow-up field study and MSNs and CTS-MSNs reduced disease progression by 12 and 15%, respectively. A significant 32% increase was observed for chlorophyll content for plants treated with CTS-MSNs. Perhaps most importantly, nanoscale silica seed treatment significantly increased (23–68%) the micronutrient (Zn, Mn, Mg, K, B) content of soybean pods, suggesting a potential sustainable strategy for nano-enabled biofortification to address nutrition insecurity. Overall, these findings indicate that MSN and CTS-MSN seed treatments in soybeans enable disease suppression and increase plant health as part of a nano-enabled strategy for sustainable agriculture. 

Speech Emotion Recognition (SER) faces a distinct challenge compared to other speech-related tasks because the annotations will show the subjective emotional perceptions of different annotators. Previous SER studies often view the subjectivity of emotion perception as noise by using the majority rule or plurality rule to obtain the consensus labels. However, these standard approaches overlook the valuable information of labels that do not agree with the consensus and make it easier for the test set. Emotion perception can have co-occurring emotions in realistic conditions, and it is unnecessary to regard the disagreement between raters as noise. To bridge the SER into a multi-label task, we introduced an “all-inclusive rule,” which considers all available data, ratings, and distributional labels as multi-label targets and a complete test set. We demonstrated that models trained with multi-label targets generated by the proposed AR outperform conventional single-label methods across incomplete and complete test sets. 

Audio-visual emotion recognition (AVER) has been an important research area in human-computer interaction (HCI). Traditionally, audio-visual emotional datasets and corresponding models derive their ground truths from annotations obtained by raters after watching the audio-visual stimuli. This conventional method, however, neglects the nuanced human perception of emotional states, which varies when annotations are made under different emotional stimuli conditions—whether through unimodal or multimodal stimuli. This study investigates the potential for enhanced AVER system performance by integrating diverse levels of annotation stimuli, reflective of varying perceptual evaluations. We propose a two-stage training method to train models with the labels elicited by audio-only, face-only, and audio-visual stimuli. Our approach utilizes different levels of annotation stimuli according to which modality is present within different layers of the model, effectively modeling annotation at the unimodal and multi-modal levels to capture the full scope of emotion perception across unimodal and multimodal contexts. We conduct the experiments and evaluate the models on the CREMA-D emotion database. The proposed methods achieved the best performances in macro-/weighted-F1 scores. Additionally, we measure the model calibration, performance bias, and fairness metrics considering the age, gender, and race of the AVER systems.