Compositional generalization is a basic mechanism in human language learning, but current neural networks lack such ability. In this paper, we conduct fundamental research for encoding compositionality in neural networks. Conventional methods use a single representation for the input sentence, making it hard to apply prior knowledge of compositionality. In contrast, our approach leverages such knowledge with two representations, one generating attention maps, and the other mapping attended input words to output symbols. We reduce the entropy in each representation to improve generalization. Our experiments demonstrate significant improvements over the conventional methods in five NLP tasks including instruction learning and machine translation. In the SCAN domain, it boosts accuracies from 14.0% to 98.8% in Jump task, and from 92.0% to 99.7% in TurnLeft task. It also beats human performance on a few-shot learning task. We hope the proposed approach can help ease future research towards human-level compositional language learning. Source code is available online 1 .

Zeptomole detector: A highly sensitive giant-magnetoresistive chip and FeCo nanoparticles can be used to linearly detect 600-4500 copies of streptavidin. Under unoptimized conditions, this system also detects human IL-6 with a sensitivity 13-times higher than that of standard ELISA techniques.

No abstract

A novel giant magnetoresistive sensor and uniform high-magnetic-moment FeCo nanoparticles (12.8 nm)-based detecting platform with minimized detecting distance was developed for rapid biomolecule quantification from body fluids. Such a system demonstrates specific, accurate, and quick detection and quantification of interleukin-6, a low-abundance protein and a potential cancer biomarker, directly in 4 muL of unprocessed human sera. This platform is expected to facilitate the identification and validation of disease biomarkers. It may eventually lead to a low-cost personal medical device for chronic disease early detection, diagnosis, and prognosis.

No abstract