Papers Read on AI

Error correction is widely used in automatic speech recognition (ASR) to post-process the generated sentence, and can further reduce the word error rate (WER). Although multiple candidates are generated by an ASR system through beam search, current error correction approaches can only correct one sentence at a time, failing to leverage the voting effect1 from multiple candidates to better detect and correct error tokens. In this work, we propose FastCorrect 2, an error correction model that takes multiple ASR candidates as input for better correction accuracy.

2021: Yichong Leng, Xu Tan, Rui Wang, Linchen Zhu, Jin Xu, Linquan Liu, Tao Qin, Xiang-Yang Li, Ed Lin, Tie-Yan Liu

https://arxiv.org/pdf/2109.14420v3.pdf

Error correction is widely used in automatic speech recognition (ASR) to post-process the generated sentence, and can further reduce the word error rate (WER). Although multiple candidates are generated by an ASR system through beam search, current error correction approaches can only correct one sentence at a time, failing to leverage the voting effect1 from multiple candidates to better detect and correct error tokens. In this work, we propose FastCorrect 2, an error correction model that takes multiple ASR candidates as input for better correction accuracy.

2021: Yichong Leng, Xu Tan, Rui Wang, Linchen Zhu, Jin Xu, Linquan Liu, Tao Qin, Xiang-Yang Li, Ed Lin, Tie-Yan Liu

https://arxiv.org/pdf/2109.14420v3.pdf

Error correction is widely used in automatic speech recognition (ASR) to post-process the generated sentence, and can further reduce the word error rate (WER). Although multiple candidates are generated by an ASR system through beam search, current error correction approaches can only correct one sentence at a time, failing to leverage the voting effect1 from multiple candidates to better detect and correct error tokens. In this work, we propose FastCorrect 2, an error correction model that takes multiple ASR candidates as input for better correction accuracy.

2021: Yichong Leng, Xu Tan, Rui Wang, Linchen Zhu, Jin Xu, Linquan Liu, Tao Qin, Xiang-Yang Li, Ed Lin, Tie-Yan Liu

https://arxiv.org/pdf/2109.14420v3.pdf

We present a frame interpolation algorithm that synthesizes multiple intermediate frames from two input images with large in-between motion. Recent methods use multiple networks to estimate optical flow or depth and a separate network dedicated to frame synthesis. This is often complex and requires scarce optical flow or depth ground-truth. In this work, we present a single unified network, distinguished by a multi-scale feature extractor that shares weights at all scales, and is trainable from frames alone. 

2022: F. Reda, Janne Kontkanen, Eric Tabellion, Deqing Sun, C. Pantofaru, B. Curless

https://arxiv.org/pdf/2202.04901v1.pdf

A typical approach to programming is to first code the main execution scenario, and then focus on filling out alternative behaviors and corner cases. But, almost always, there exist unusual conditions that trigger atypical behaviors, which are hard to predict in program specifications, and are thus often not coded. In this paper, we consider the problem of detecting and recommending such missing behaviors, a task that we call code sophistication. Previous research on coding assistants usually focuses on recommending code fragments based on specifications of the intended behavior. In contrast, code sophistication happens in the absence of a specification, aiming to help developers complete the logic of their programs with missing and unspecified behaviors. We outline the research challenges to this problem and present early results showing how program logic can be completed by leveraging code structure and information about the usage of input parameters.

2022: Jessie Galasso, Michalis Famelis, H. Sahraoui

https://arxiv.org/pdf/2201.07674v1.pdf

Evolutionary computation has been shown to be a highly effective method for training neural networks, particularly when employed at scale on CPU clusters. Recent work have also showcased their effectiveness on hardware accelerators, such as GPUs, but so far such demonstrations are tailored for very specific tasks, limiting applicability to other domains. We present EvoJAX, a scalable, general purpose, hardware-accelerated neuroevolution toolkit.

2022: Yujin Tang, Yingtao Tian, David Ha

https://arxiv.org/pdf/2202.05008v1.pdf

Evolutionary computation has been shown to be a highly effective method for training neural networks, particularly when employed at scale on CPU clusters. Recent work have also showcased their effectiveness on hardware accelerators, such as GPUs, but so far such demonstrations are tailored for very specific tasks, limiting applicability to other domains. We present EvoJAX, a scalable, general purpose, hardware-accelerated neuroevolution toolkit.

2022: Yujin Tang, Yingtao Tian, David Ha

https://arxiv.org/pdf/2202.05008v1.pdf

Fine-tuning reinforcement learning (RL) models has been challenging because of a lack of large scale off-the-shelf datasets as well as high variance in transferability among different environments. Recent work has looked at tackling offline RL from the perspective of sequence modeling with improved results as result of the introduction of the Transformer architecture. However, when the model is trained from scratch, it suffers from slow convergence speeds. In this paper, we look to take advantage of this formulation of reinforcement learning as sequence modeling and investigate the transferability of pre-trained sequence models on other domains (vision, language) when finetuned on offline RL tasks (control, games). To this end, we also propose techniques to improve transfer between these domains. Results show consistent performance gains in terms of both convergence speed and reward on a variety of environments, accelerating training by 3-6x and achieving state-of-the-art performance in a variety of tasks using Wikipedia-pretrained and GPT2 language models. We hope that this work not only brings light to the potentials of leveraging generic sequence modeling techniques and pre-trained models for RL, but also inspires future work on sharing knowledge between generative modeling tasks of completely different domains.

2022: Machel Reid, Yutaro Yamada, S. Gu

https://arxiv.org/pdf/2201.12122v1.pdf

We present LSeg, a novel model for language-driven semantic image segmentation. LSeg uses a text encoder to compute embeddings of descriptive input labels (e.g., “grass” or “building”) together with a transformer-based image encoder that computes dense per-pixel embeddings of the input image. The image encoder is trained with a contrastive objective to align pixel embeddings to the text embedding of the corresponding semantic class. The text embeddings provide a flexible label representation in which semantically similar labels map to similar regions in the embedding space.

2022: Boyi Li, Kilian Q. Weinberger, Serge J. Belongie, V. Koltun, René Ranftl Ranked

#2 on Few-Shot Semantic Segmentation on FSS-1000

https://arxiv.org/pdf/2201.03546v1.pdf

Dropout is a powerful and widely used technique to regularize the training of deep neural networks. Though effective and performing well, the randomness introduced by dropout causes unnegligible inconsistency between training and inference. In this paper, we introduce a simple consistency training strategy to regularize dropout, namely R-Drop, which forces the output distributions of different sub models generated by dropout to be consistent with each other.

2021: Xiaobo Liang, Lijun Wu, Juntao Li, Yue Wang, Qi Meng, Tao Qin, Wei Chen, Min Zhang, Tie-Yan Liu Ranked

#2 on Abstractive Text Summarization on CNN / Daily Mail

https://arxiv.org/pdf/2106.14448v2.pdf