Social media is daily creating massive multimedia content with paired image and text, presenting the pressing need to automate the vision and language understanding for various multimodal classification tasks. Compared to the commonly researched visual-lingual data, social media posts tend to exhibit more implicit image-text relations. To better glue the cross-modal semantics therein, we capture hinting features from user comments, which are retrieved via jointly leveraging visual and lingual similarity. Afterwards, the classification tasks are explored via self-training in a teacher-student framework, motivated by the usually limited labeled data scales in existing benchmarks. Substantial experiments are conducted on four multimodal social media benchmarks for image text relation classification, sarcasm detection, sentiment classification, and hate speech detection. The results show that our method further advances the performance of previous state-of-the-art models, which do not employ comment modeling or self-training.

Recurrent neural networks are effective models to process sequences. However, they are unable to learn long-term dependencies because of their inherent sequential nature. As a solution, Vaswani et al. introduced the Transformer, a model solely based on the attention mechanism that is able to relate any two positions of the input sequence, hence modelling arbitrary long dependencies. The Transformer has improved the state-of-the-art across numerous sequence modelling tasks. However, its effectiveness comes at the expense of a quadratic computational and memory complexity with respect to the sequence length, hindering its adoption. Fortunately, the deep learning community has always been interested in improving the models' efficiency, leading to a plethora of solutions such as parameter sharing, pruning, mixed-precision, and knowledge distillation. Recently, researchers have directly addressed the Transformer's limitation by designing lower-complexity alternatives such as the Longformer, Reformer, Linformer, and Performer. However, due to the wide range of solutions, it has become challenging for researchers and practitioners to determine which methods to apply in practice in order to meet the desired trade-off between capacity, computation, and memory. This survey addresses this issue by investigating popular approaches to make Transformers faster and lighter and by providing a comprehensive explanation of the methods' strengths, limitations, and underlying assumptions.

Private thoughts may not be private for much longer, heralding a nightmarish world where political views, thoughts, stray obsessions and feelings could be interrogated and punished all thanks to advances in neurotechnology. Or at least that is what one of the world's leading brain scientists believes. In a new book, The Battle for Your Brain, Duke University bioscience professor Nita Farahany argues that such intrusions into the human mind by technology are so close that a public discussion is long overdue and lawmakers should immediately establish brain protections as it would for any other area of personal liberty. Advances in hacking and tracking thoughts, with Orwellian fears of mind control running just below the surface, is the subject of Farahany's scholarship alongside urgent calls for legislative guarantees to thought privacy, including freedoms from "cognitive fingerprinting", that lie within an area of ethics broadly termed "cognitive liberty". Certainly the field is advancing rapidly.

The synthesis of tactile sensing with compliance is essential to many fields, from agricultural usages like fruit picking, to sustainability practices such as sorting recycling, to the creation of safe home-care robots for the elderly to age with dignity. From tactile sensing, we can discern material properties, recognize textures, and determine softness, while with compliance, we are able to securely and safely interact with the objects and the environment around us. These two abilities can culminate into a useful soft robotic gripper, such as the original GelSight Fin Ray, which is able to grasp a large variety of different objects and also perform a simple household manipulation task: wine glass reorientation. Although the original GelSight Fin Ray solves the problem of interfacing a generally rigid, high-resolution sensor with a soft, compliant structure, we can improve the robustness of the sensor and implement techniques that make such camera-based tactile sensors applicable to a wider variety of soft robot designs. We first integrate flexible mirrors and incorporate the rigid electronic components into the base of the gripper, which greatly improves the compliance of the Fin Ray structure. Then, we synthesize a flexible and high-elongation silicone adhesive-based fluorescent paint, which can provide good quality 2D tactile localization results for our sensor. Finally, we incorporate all of these techniques into a new design: the Baby Fin Ray, which we use to dig through clutter, and perform successful classification of nuts in their shells. The supplementary video can be found here: https://youtu.be/_oD_QFtYTPM

Adapting generic speech recognition models to specific individuals is a challenging problem due to the scarcity of personalized data. Recent works have proposed boosting the amount of training data using personalized text-to-speech synthesis. Here, we ask two fundamental questions about this strategy: when is synthetic data effective for personalization, and why is it effective in those cases? To address the first question, we adapt a state-of-the-art automatic speech recognition (ASR) model to target speakers from four benchmark datasets representative of different speaker types. We show that ASR personalization with synthetic data is effective in all cases, but particularly when (i) the target speaker is underrepresented in the global data, and (ii) the capacity of the global model is limited. To address the second question of why personalized synthetic data is effective, we use controllable speech synthesis to generate speech with varied styles and content. Surprisingly, we find that the text content of the synthetic data, rather than style, is important for speaker adaptation. These results lead us to propose a data selection strategy for ASR personalization based on speech content.

Suppose we are given an $n$-dimensional order-3 symmetric tensor $T \in (\mathbb{R}^n)^{\otimes 3}$ that is the sum of $r$ random rank-1 terms. The problem of recovering the rank-1 components is possible in principle when $r \lesssim n^2$ but polynomial-time algorithms are only known in the regime $r \ll n^{3/2}$. Similar "statistical-computational gaps" occur in many high-dimensional inference tasks, and in recent years there has been a flurry of work on explaining the apparent computational hardness in these problems by proving lower bounds against restricted (yet powerful) models of computation such as statistical queries (SQ), sum-of-squares (SoS), and low-degree polynomials (LDP). However, no such prior work exists for tensor decomposition, largely because its hardness does not appear to be explained by a "planted versus null" testing problem. We consider a model for random order-3 tensor decomposition where one component is slightly larger in norm than the rest (to break symmetry), and the components are drawn uniformly from the hypercube. We resolve the computational complexity in the LDP model: $O(\log n)$-degree polynomial functions of the tensor entries can accurately estimate the largest component when $r \ll n^{3/2}$ but fail to do so when $r \gg n^{3/2}$. This provides rigorous evidence suggesting that the best known algorithms for tensor decomposition cannot be improved, at least by known approaches. A natural extension of the result holds for tensors of any fixed order $k \ge 3$, in which case the LDP threshold is $r \sim n^{k/2}$.

Meetings are increasingly important for collaborations. Action items in meeting transcripts are crucial for managing post-meeting to-do tasks, which usually are summarized laboriously. The Action Item Detection task aims to automatically detect meeting content associated with action items. However, datasets manually annotated with action item detection labels are scarce and in small scale. We construct and release the first Chinese meeting corpus with manual action item annotations. In addition, we propose a Context-Drop approach to utilize both local and global contexts by contrastive learning, and achieve better accuracy and robustness for action item detection. We also propose a Lightweight Model Ensemble method to exploit different pre-trained models. Experimental results on our Chinese meeting corpus and the English AMI corpus demonstrate the effectiveness of the proposed approaches.

Counterfactual explanation is one branch of interpretable machine learning that produces a perturbation sample to change the model's original decision. The generated samples can act as a recommendation for end-users to achieve their desired outputs. Most of the current counterfactual explanation approaches are the gradient-based method, which can only optimize the differentiable loss functions with continuous variables. Accordingly, the gradient-free methods are proposed to handle the categorical variables, which however have several major limitations: 1) causal relationships among features are typically ignored when generating the counterfactuals, possibly resulting in impractical guidelines for decision-makers; 2) the counterfactual explanation algorithm requires a great deal of effort into parameter tuning for dertermining the optimal weight for each loss functions which must be conducted repeatedly for different datasets and settings. In this work, to address the above limitations, we propose a prototype-based counterfactual explanation framework (ProCE). ProCE is capable of preserving the causal relationship underlying the features of the counterfactual data. In addition, we design a novel gradient-free optimization based on the multi-objective genetic algorithm that generates the counterfactual explanations for the mixed-type of continuous and categorical features. Numerical experiments demonstrate that our method compares favorably with state-of-the-art methods and therefore is applicable to existing prediction models. All the source codes and data are available at \url{https://github.com/tridungduong16/multiobj-scm-cf}.

Technology can be a topic difficult to understand and make predictions on, even for those with a strong technical background and perceived expertise. From Ethernet's creator Robert Metcalfe's 1995 prediction that the internet would "catastrophically collapse" by the next year to Intel's prediction that 3D TV was the future, it is clear that predicting tech trends is a difficult endeavor. No matter how hard predicting the future of technology is, every new technology that creates disruption will go through this cycle. Most recently, we have gone through multiple hype cycles with innovations like blockchain, cryptocurrency, the metaverse, VR, and now, AI. Every single of these technologies has captivated not only the public but also developers and investors, blurring the line between facts and fiction.

We introduce Deep Augmentation, an approach to data augmentation using dropout to dynamically transform a targeted layer within a neural network, with the option to use the stop-gradient operation, offering significant improvements in model performance and generalization. We demonstrate the efficacy of Deep Augmentation through extensive experiments on contrastive learning tasks in computer vision and NLP domains, where we observe substantial performance gains with ResNets and Transformers as the underlying models. Our experimentation reveals that targeting deeper layers with Deep Augmentation outperforms augmenting the input data, and the simple network- and data-agnostic nature of this approach enables its seamless integration into computer vision and NLP pipelines.