Functional Geometry Guided Protein Sequence and Backbone Structure Co-Design

Proteins are macromolecules responsible for essential functions in almost all living organisms. Designing reasonable proteins with desired functions is crucial. A protein's sequence and structure are strongly correlated and they together determine its function. In this paper, we propose NAEPro, a model to jointly design Protein sequence and structure based on automatically detected functional sites. NAEPro is powered by an interleaving network of attention and equivariant layers, which can capture global correlation in a whole sequence and local influence from nearest amino acids in three dimensional (3D) space. Such an architecture facilitates effective yet economic message passing at two levels. We evaluate our model and several strong baselines on two protein datasets, $\beta$-lactamase and myoglobin. Experimental results show that our model consistently achieves the highest amino acid recovery rate, TM-score, and the lowest RMSD among all competitors. These findings prove the capability of our model to design protein sequences and structures that closely resemble their natural counterparts. Furthermore, in-depth analysis further confirms our model's ability to generate highly effective proteins capable of binding to their target metallocofactors. We provide code, data and models in Github.

Joint Design of Protein Sequence and Structure based on Motifs

Designing novel proteins with desired functions is crucial in biology and chemistry. However, most existing work focus on protein sequence design, leaving protein sequence and structure co-design underexplored. In this paper, we propose GeoPro, a method to design protein backbone structure and sequence jointly. Our motivation is that protein sequence and its backbone structure constrain each other, and thus joint design of both can not only avoid nonfolding and misfolding but also produce more diverse candidates with desired functions. To this end, GeoPro is powered by an equivariant encoder for three-dimensional (3D) backbone structure and a protein sequence decoder guided by 3D geometry. Experimental results on two biologically significant metalloprotein datasets, including $\beta$-lactamases and myoglobins, show that our proposed GeoPro outperforms several strong baselines on most metrics. Remarkably, our method discovers novel $\beta$-lactamases and myoglobins which are not present in protein data bank (PDB) and UniProt. These proteins exhibit stable folding and active site environments reminiscent of those of natural proteins, demonstrating their excellent potential to be biologically functional.

Adversarial Object Rearrangement in Constrained Environments with Heterogeneous Graph Neural Networks

Adversarial object rearrangement in the real world (e.g., previously unseen or oversized items in kitchens and stores) could benefit from understanding task scenes, which inherently entail heterogeneous components such as current objects, goal objects, and environmental constraints. The semantic relationships among these components are distinct from each other and crucial for multi-skilled robots to perform efficiently in everyday scenarios. We propose a hierarchical robotic manipulation system that learns the underlying relationships and maximizes the collaborative power of its diverse skills (e.g., pick-place, push) for rearranging adversarial objects in constrained environments. The high-level coordinator employs a heterogeneous graph neural network (HetGNN), which reasons about the current objects, goal objects, and environmental constraints; the low-level 3D Convolutional Neural Network-based actors execute the action primitives. Our approach is trained entirely in simulation, and achieved an average success rate of 87.88% and a planning cost of 12.82 in real-world experiments, surpassing all baseline methods. Supplementary material is available at https://sites.google.com/umn.edu/versatile-rearrangement.

Disentangled Information Bottleneck guided Privacy-Protective JSCC for Image Transmission

Joint source and channel coding (JSCC) has attracted increasing attention due to its robustness and high efficiency. However, JSCC is vulnerable to privacy leakage due to the high relevance between the source image and channel input. In this paper, we propose a disentangled information bottleneck guided privacy-protective JSCC (DIB-PPJSCC) for image transmission, which aims at protecting private information as well as achieving superior communication performance at the legitimate receiver. In particular, we propose a DIB objective to disentangle private and public information. The goal is to compress the private information in the public subcodewords, preserve the private information in the private subcodewords and improve the reconstruction quality simultaneously. In order to optimize JSCC neural networks using the DIB objective, we derive a differentiable estimation of the DIB objective based on the variational approximation and the density-ratio trick. Additionally, we design a password-based privacy-protective (PP) algorithm which can be jointly optimized with JSCC neural networks to encrypt the private subcodewords. Specifically, we employ a private information encryptor to encrypt the private subcodewords before transmission, and a corresponding decryptor to recover the private information at the legitimate receiver. A loss function for jointly training the encryptor, decryptor and JSCC decoder is derived based on the maximum entropy principle, which aims at maximizing the eavesdropping uncertainty as well as improving the reconstruction quality. Experimental results show that DIB-PPJSCC can reduce the eavesdropping accuracy on private information up to $15\%$ and reduce $10\%$ inference time compared to existing privacy-protective JSCC and traditional separate methods.

How to Generate Popular Post Headlines on Social Media?

Posts, as important containers of user-generated-content pieces on social media, are of tremendous social influence and commercial value. As an integral components of a post, the headline has a decisive contribution to the post's popularity. However, current mainstream method for headline generation is still manually writing, which is unstable and requires extensive human effort. This drives us to explore a novel research question: Can we automate the generation of popular headlines on social media? We collect more than 1 million posts of 42,447 celebrities from public data of Xiaohongshu, which is a well-known social media platform in China. We then conduct careful observations on the headlines of these posts. Observation results demonstrate that trends and personal styles are widespread in headlines on social medias and have significant contribution to posts's popularity. Motivated by these insights, we present MEBART, which combines Multiple preference-Extractors with Bidirectional and Auto-Regressive Transformers (BART), capturing trends and personal styles to generate popular headlines on social medias. We perform extensive experiments on real-world datasets and achieve state-of-the-art performance compared with several advanced baselines. In addition, ablation and case studies demonstrate that MEBART advances in capturing trends and personal styles.

Learning to Generate Lumped Hydrological Models

In a lumped hydrological model structure, the hydrological function of a catchment is characterized by only a few parameters. Given a set of parameter values, a numerical function useful for hydrological prediction is generated. Thus, this study assumes that the hydrological function of a catchment can be sufficiently well characterized by a small number of latent variables. By specifying the variable values, a numerical function resembling the hydrological function of a real-world catchment can be generated using a generative model. In this study, a deep learning method is used to learn both the generative model and the latent variable values of different catchments directly from their climate forcing and runoff data, without using catchment attributes. The generative models can be used similarly to a lumped model structure, i.e., by estimating the optimal parameter or latent variable values using a generic model calibration algorithm, an optimal numerical model can be derived. In this study, generative models using eight latent variables were learned from data from over 3,000 catchments worldwide, and the learned generative models were applied to model over 700 different catchments using a generic calibration algorithm. The quality of the resulting optimal models was generally comparable to or better than that obtained using 36 different types of lump model structures or using non-generative deep learning methods. In summary, this study presents a data-driven approach for representing the hydrological function of a catchment in low-dimensional space and a method for reconstructing specific hydrological functions from the representations.

Privacy-Aware Joint Source-Channel Coding for image transmission based on Disentangled Information Bottleneck

Current privacy-aware joint source-channel coding (JSCC) works aim at avoiding private information transmission by adversarially training the JSCC encoder and decoder under specific signal-to-noise ratios (SNRs) of eavesdroppers. However, these approaches incur additional computational and storage requirements as multiple neural networks must be trained for various eavesdroppers' SNRs to determine the transmitted information. To overcome this challenge, we propose a novel privacy-aware JSCC for image transmission based on disentangled information bottleneck (DIB-PAJSCC). In particular, we derive a novel disentangled information bottleneck objective to disentangle private and public information. Given the separate information, the transmitter can transmit only public information to the receiver while minimizing reconstruction distortion. Since DIB-PAJSCC transmits only public information regardless of the eavesdroppers' SNRs, it can eliminate additional training adapted to eavesdroppers' SNRs. Experimental results show that DIB-PAJSCC can reduce the eavesdropping accuracy on private information by up to 20\% compared to existing methods.

NICE: CVPR 2023 Challenge on Zero-shot Image Captioning

In this report, we introduce NICE (New frontiers for zero-shot Image Captioning Evaluation) project and share the results and outcomes of 2023 challenge. This project is designed to challenge the computer vision community to develop robust image captioning models that advance the state-of-the-art both in terms of accuracy and fairness. Through the challenge, the image captioning models were tested using a new evaluation dataset that includes a large variety of visual concepts from many domains. There was no specific training data provided for the challenge, and therefore the challenge entries were required to adapt to new types of image descriptions that had not been seen during training. This report includes information on the newly proposed NICE dataset, evaluation methods, challenge results, and technical details of top-ranking entries. We expect that the outcomes of the challenge will contribute to the improvement of AI models on various vision-language tasks.

Cross-Utterance Conditioned VAE for Speech Generation

Speech synthesis systems powered by neural networks hold promise for multimedia production, but frequently face issues with producing expressive speech and seamless editing. In response, we present the Cross-Utterance Conditioned Variational Autoencoder speech synthesis (CUC-VAE S2) framework to enhance prosody and ensure natural speech generation. This framework leverages the powerful representational capabilities of pre-trained language models and the re-expression abilities of variational autoencoders (VAEs). The core component of the CUC-VAE S2 framework is the cross-utterance CVAE, which extracts acoustic, speaker, and textual features from surrounding sentences to generate context-sensitive prosodic features, more accurately emulating human prosody generation. We further propose two practical algorithms tailored for distinct speech synthesis applications: CUC-VAE TTS for text-to-speech and CUC-VAE SE for speech editing. The CUC-VAE TTS is a direct application of the framework, designed to generate audio with contextual prosody derived from surrounding texts. On the other hand, the CUC-VAE SE algorithm leverages real mel spectrogram sampling conditioned on contextual information, producing audio that closely mirrors real sound and thereby facilitating flexible speech editing based on text such as deletion, insertion, and replacement. Experimental results on the LibriTTS datasets demonstrate that our proposed models significantly enhance speech synthesis and editing, producing more natural and expressive speech.

NICE 2023 Zero-shot Image Captioning Challenge

In this report, we introduce NICE project\footnote{\url{https://nice.lgresearch.ai/}} and share the results and outcomes of NICE challenge 2023. This project is designed to challenge the computer vision community to develop robust image captioning models that advance the state-of-the-art both in terms of accuracy and fairness. Through the challenge, the image captioning models were tested using a new evaluation dataset that includes a large variety of visual concepts from many domains. There was no specific training data provided for the challenge, and therefore the challenge entries were required to adapt to new types of image descriptions that had not been seen during training. This report includes information on the newly proposed NICE dataset, evaluation methods, challenge results, and technical details of top-ranking entries. We expect that the outcomes of the challenge will contribute to the improvement of AI models on various vision-language tasks.