Konka (China)

In the Knowledge Grounded Dialogue (KGD) generation, the explicit modeling of instance-variety of knowledge specificity and its seamless fusion with the dialogue context remains challenging. This paper presents an innovative approach, the Knowledge Interpolated conditional Variational auto-encoder (KIV), to address these issues. In particular, KIV introduces a novel interpolation mechanism to fuse two latent variables: independently encoding dialogue context and grounded knowledge. This distinct fusion of context and knowledge in the semantic space enables the interpolated latent variable to guide the decoder toward generating more contextually rich and engaging responses. We further explore deterministic and probabilistic methodologies to ascertain the interpolation weight, capturing the level of knowledge specificity. Comprehensive empirical analysis conducted on the Wizard-of-Wikipedia and Holl-E datasets verifies that the responses generated by our model performs better than strong baselines, with notable performance improvements observed in both automatic metrics and manual evaluation.

The accurate recognition of emotions in conversations helps understand the speaker’s intentions and facilitates various analyses in artificial intelligence, especially in human–computer interaction systems. However, most previous methods need more ability to track the different emotional states of each speaker in a dialogue. To alleviate this dilemma, we propose a new approach, Multi-Task Learning and Multi-Fusion AudioText Emotion Recognition in Conversation (MMATERIC) for emotion recognition in conversation. MMATERIC can refer to and combine the benefits of two distinct tasks: emotion recognition in text and emotion recognition in speech, and production of fused multimodal features to recognize the emotions of different speakers in dialogue. At the core of MATTERIC are three modules: an encoder with multimodal attention, a speaker emotion detection unit (SED-Unit), and a decoder with speaker emotion detection Bi-LSTM (SED-Bi-LSTM). Together, these three modules model the changing emotions of a speaker at a given moment in a conversation. Meanwhile, we adopt multiple fusion strategies in different stages, mainly using model fusion and decision stage fusion to improve the model’s accuracy. Simultaneously, our multimodal framework allows features to interact across modalities and allows potential adaptation flows from one modality to another. Our experimental results on two benchmark datasets show that our proposed method is effective and outperforms the state-of-the-art baseline methods. The performance improvement of our method is mainly attributed to the combination of three core modules of MATTERIC and the different fusion methods we adopt in each stage.

Point-pair registration in a real scene remains a challenging task, due to the complexity of solving three transformations (scale, rotation, and displacement) simultaneously, and the influence of noise and outliers. Aimed at this problem, a registration algorithm based on histogram and vector operations is proposed in this paper. This approach converts point-based operations into vector-based operations, thereby decomposing the registration process into three independent steps solving for scale transformation factors, rotation matrices, and displacement vectors, which reduces the complexity of the solution and avoids the effects of scaling in the other two processes. The influence of outliers on the global transformation matrix is simultaneously eliminated using a histogram-based approach. Algorithm performance was evaluated through a comparison with the most commonly used SVD method in a series of validation experiments, with results showing that our methodology was superior to SVD in the cases with scaling transformation or outliers.

Principal Component Analysis (PCA) is the most widely used unsupervised subspace learning method, and lots of its variants have been developed. With so many proposed PCA-like methods, it is still not clear that which features are better or worse for principal components, especially when the data suffers from outliers. To this end, we propose Robust Principal Component Analysis via joint l 2,1 -norms minimization, which provides new insights into two crucial issues of PCA: feature selection and robustness to outliers. Unlike other PCA-like methods, the proposed method is able to select effective features for reconstruction by using the l 2,1 -norm regularization term. More specific, we first use a l 2,1 -norm based transformation matrix to select effective features that can effectively characterize key components (e.g., the eyes and the nose in a face image), and then use an orthogonal transformation matrix to recover the original data from the selected data representation. In this way, the key components can be well recovered by using the effective features selected by a learned transformation matrix. On the other hand, we also impose l 2,1 -norm on a loss term to select clean samples to recover its same class samples but with outliers. A simple yet effective optimization algorithm is proposed to solve the resulting optimization problem. Experiments on six datasets demonstrate the effectiveness of the proposed method.

Visual tracking is a challenging problem in computer vision. Most state-of-the-art visual trackers are lack of an ability to adapt affine transforms of objects. The Lucas-Kanade(L-K) algorithm is a classic gradient decent search method which has been widely applied in template matching, image alignment and visual tracking. Despite the L-K algorithm has a good property on adapting targets affine transforms, it always falls into a local optimal since the background information is accumulatively introduced to the template. In this paper, we propose a novel approach, called the Re-located L-K algorithm(RL-K), to handle affine transforms of objects. In particular, we utilize the Multiple Channel Correlation Filters(MMCFs) to deal with the translations of the object before the L-K's gradient decent search step. In addition, we integrate the gradient orientation features into the L-K algorithm to improve its robustness to visual tracking. The experimental results show that, comparing with the state-of-the-art affine tracking approaches, our approach is running at real-time without accuracy losing.

Lack or downregulation of dopamine D2 receptor (D2R) function in mice increases the vulnerability to renal inflammation. Recent studies from our laboratory showed that long-term (28 days) selective down regulation of D2R in one kidney in mice with two intact kidneys increases systolic blood pressure and the expression of inflammatory and injury markers. We hypothesized that rescuing D2R function by re-expressing the receptor in the kidney of mice with D2R silencing will reduce renal inflammation and injury, and normalize blood pressure. To test our hypothesis we silenced D2R expression by the renal subcapsular infusion of D2RsiRNA and then we rescued the silenced D2R expression using retrograde ureteral infusion of AAV9 vector carrying wild-type D2R (D2RAAV). Mice were treated with non-silencing siRNA (NS siRNA) or D2RsiRNA (3 μg/day, n=6 per group) using 28 day osmotic minipumps. Fourteen days following the initiation of siRNA treatment, each group was treated with control AAV (CAAV) or D2RAAV (n=3 per group, 1e+11 viral genome particles per mouse). Fourteen days following AAV administration, blood pressure was recorded and organs were collected. D2R expression was decreased in D2RsiRNA-treated kidneys compared with NS siRNA-treated kidneys (immunoblotting: 54 ± 0.8 vs 100± 22 %; P<0.05). D2RAAV treatment increased D2R expression (7.5 -11 fold; P<0.01) in both D2R and NS siRNA-treated mice. Mice treated with D2RsiRNA+CAAV had increased systolic blood pressure (121±3 mmHg; P<0.05) in comparison with mice treated with D2RsiRNA+D2RAAV (100±6 mm Hg) (rescued mice), NS siRNA +CAAV (101±4 mm Hg), or NS siRNA +D2RAAV (101±1 mm Hg). The expression of TGF-β was higher in the mice treated with D2RsiRNA+CAAV than in D2R rescued mice (D2RsiRNA+D2RAAV) (100±22 vs 54± 11%; P<0.05). The number of cells positive for the marker of proliferation ki-67 was higher in kidneys of D2RsiRNA+CAAV mice than in D2RsiRNA+D2RAAV (32± 5 vs 15±2 per field; P<0.01). In conclusion, these results demonstrate that increased blood pressure and renal injury due to D2R silencing could be rescued by over expression of D2R using AAV9 vector in the kidney. Furthermore, these data confirms the protective effects of renal D2R and may provide the basis for designing novel therapies for kidney disease.

Human gene variants account for only a small proportion of essential hypertension in humans and have not been shown to produce hypertension in transgenic mice. The association between hypertension and 11 allelic variants of 8 candidate genes in humans (n=1074) was studied. We also evaluated the response to angiotensin receptor blockers (ARBs) in humans (n=681) and transgenic mice generated to express the human wild-type G protein-coupled receptor kinase 4γ ( hGRK4 γ WT ) or the 142V allele (hGRK4γ 142V ). hGRK4 γ variants ( R65L , A142V , and A486V ) were associated with essential hypertension; hGRK4γ 142V , by itself, predicted the hypertensive phenotype (60% accuracy); hGRK4γ 142V also interacted with GNB3 and PAI-1 . Human hypertensive carriers of hGRK4 γ 142V (n=68) had a greater decrease in systolic blood pressure (SBP mm Hg, 19.36, vs. 14.58, P=0.001) in response to ARBs than non-carrier (n=135) hypertensives. hGRK4γ WT transgenic mice (anesthetized, n=42; conscious [n=12, day=118±0.8, night=131±1]) were normotensive while hGRK4γ 142V mice (anesthetized, n=54;conscious [n=12, day=128±2, night=148±3]) had hypertension, increased renal AT 1 R expression due to decreased histone deacetylase 1 (HDAC1) activity, a greater pressor response to angiotensin II (1000 ng/kg/minX30’)(n=6-9/group, WT=Δ+14±4%, 142V=Δ+26±3%), and a greater fall in SBP with candesartan (2mg/kg/dayX4d) (n=7/group, WT=Δ-5±2%,142V=Δ-29±3%); deletion of Agtr1a normalized the high BP of hGRK4γ 142V mice (n=5-13/group). To determine the role of renal HDAC1 on BP, HDAC1 was silenced with HDAC-siRNA. Renal-specific HDAC1, but not HDAC2, silencing increased SBP (HDAC1-siRNA=112±5; mock=91±2; P<0.004, n=5/group). The BP response to ARBs was greater in HDAC1-siRNA- than mock-treated mice. These studies show that hGRK4γ 142V is associated with human essential hypertension, produces AT 1 R-sensitive hypertension in hGRK4γ 142V transgenic mice, and predicts a greater decrease in BP in response to ARBs in humans with essential hypertension and hGRK4γ 142V transgenic mice. This is the first demonstration of a concordance between rodents and humans for genetics, signaling pathway, and pharmacogenetics of essential hypertension.

Tasa de infeccion de la pudricion del tallo en maiz causada por Fusarium moniliforme. El objetivo de este trabajo fue determinar la resistencia horizontal de 134 lineas S1 de maiz al hongo Fusarium moniliforme. La enfermedad fue inducida mediante inoculacion en la etapa de floracion masculina bajo condiciones de campo utilizando la tecnica del palillo. Las lineas se evaluaron en Mexico en 2002. La estimacion de la enfermedad se hizo en dos ocasiones con una separacion de 20 dias. Las evaluaciones mostraron que las lineas Roque-801, 804, 1103 y 1405 fueron resistentes en la primera evaluacion. Las tasas de incremento de la enfermedad variaron de 0,01 a 0,229 por unidad por dia. Las lineas mostraron bajos niveles de enfermedad de 0 a 1 % en la primera evaluacion. En la segunda evaluacion, los niveles de enfermedad variaron de 49 a 100 %. En un programa de mejoramiento genetico estas lineas pueden ser utilizadas para mejorar la resistencia horizontal del maiz a la pudricion del tallo causada por Fusarium moniliforme.