Multi-Hop Question Generation Using Hierarchical Encoding-Decoding and Context Switch Mechanism

Neural auto-regressive sequence-to-sequence models have been dominant in text generation tasks, especially the question generation task. However, neural generation models suffer from the global and local semantic semantic drift problems. Hence, we propose the hierarchical encoding–decoding mechanism that aims at encoding rich structure information of the input passages and reducing the variance in the decoding phase. In the encoder, we hierarchically encode the input passages according to its structure at four granularity-levels: [word, chunk, sentence, document]-level. Second, we progressively select the context vector from the document-level representations to the word-level representations at each decoding time step. At each time-step in the decoding phase, we progressively select the context vector from the document-level representations to word-level. We also propose the context switch mechanism that enables the decoder to use the context vector from the last step when generating the current word at each time-step.It provides a means of improving the stability of the text generation process during the decoding phase when generating a set of consecutive words. Additionally, we inject syntactic parsing knowledge to enrich the word representations. Experimental results show that our proposed model substantially improves the performance and outperforms previous baselines according to both automatic and human evaluation. Besides, we implement a deep and comprehensive analysis of generated questions based on their types.

Tuning in scene-preferring cortex for mid-level visual features gives rise to selectivity across multiple levels of stimulus complexity

The scene-preferring portion of the human ventral visual stream, known as the parahippocampal place area (PPA), responds to scenes and landmark objects, which tend to be large in real-world size, fixed in location, and inanimate. However, the PPA also exhibits preferences for low-level contour statistics, including rectilinearity and cardinal orientations, that are not directly predicted by theories of scene- and landmark-selectivity. It is unknown whether these divergent findings of both low- and high-level selectivity in the PPA can be explained by a unified computational theory. To address this issue, we fit hierarchical computational models of mid-level tuning to the image-evoked fMRI responses of the PPA, and we performed a series of high-throughput experiments on these models. Our findings show that hierarchical encoding models of the PPA exhibit emergent selectivity across multiple levels of complexity, giving rise to high-level preferences along dimensions of real-world size, fixedness, and naturalness/animacy as well as low-level preferences for rectilinear shapes and cardinal orientations. These results reconcile disparate theories of PPA function in a unified model of mid-level visual representation, and they demonstrate how multifaceted selectivity profiles naturally emerge from the hierarchical computations of visual cortex and the natural statistics of images.

Change in Evaluation Mode Can Cause a Cheerleader Effect

The cheerleader effect describes the phenomenon whereby faces are perceived as being more attractive when flanked by other faces than when they are perceived in isolation. At least four theories predict the cheerleader effect. Two visual memory processes could cause a cheerleader effect. First, visual information will sometimes be averaged in the visual memory: the averaging of faces could increase the perceived attractiveness of all the faces flanked by other faces. Second, information will often be combined into a higher-order concept. This hierarchical encoding suggests that information processing causes faces to appear more attractive when flanked by highly attractive faces. Two further explanations posit that comparison processes cause the cheerleader effect. While contrast effects predict that a difference between the target face and the flanking faces causes the cheerleader effect due to comparison processes, a change in the evaluation mode, which alters the standard of comparison between joint and separate evaluation of faces, could be sufficient for producing a cheerleader effect. This leads to the prediction that even when there is no contrast between the attractiveness of the target face and the flanking faces, a cheerleader effect could occur. The results of one experiment support this prediction. The findings of this study have practical implications, such as for individuals who post selfies on social media. An individual’s face will appear more attractive in a selfie taken with people of low attractiveness than in a selfie without other people, even when all the faces have equally low levels of attractiveness.

Attention differentially affects acoustic and phonetic feature encoding in a multispeaker environment

Humans have the remarkable ability to selectively focus on a single talker in the midst of other competing talkers. The neural mechanisms that underlie this phenomenon remain incompletely understood. In particular, there has been longstanding debate over whether attention operates at an early or late stage in the speech processing hierarchy. One way to better understand this is to examine how attention might differentially affect neurophysiological indices of hierarchical acoustic and linguistic speech representations. In this study, we do this by using encoding models to identify neural correlates of speech processing at various levels of representation. Specifically, using EEG recorded during a “cocktail party” attention experiment, we show that phonetic feature processing is evident for attended, but not unattended speech. Furthermore, we show that attention specifically enhances isolated indices of phonetic feature processing, but that such attention effects are not apparent for isolated measures of acoustic processing. These results provide new insights into the effects of attention on different pre-lexical representations of speech, insights that complement recent anatomical accounts of the hierarchical encoding of attended speech. Furthermore, our findings support the notion that – for attended speech – phonetic features are processed as a distinct stage, separate from the processing of the speech acoustics.

Core–shell metal–organic frameworks and hierarchical host–guest structures toward water-stable luminescence of lanthanide complexes in encoding beads

Hetero-MOF domains in hybrid core–shell nanoparticles realize encapsulation and protection of lanthanide complexes for water-stable photoluminescence, facilitating hierarchical encoding beads with a 3D host–guest structure.