Abstract:Streaming automatic speech recognition (ASR) is very important for many real-world ASR applications. However, a notable challenge for streaming ASR systems lies in balancing operational performance against latency constraint. Recently, a method of chunking, simulating future context and decoding, called CUSIDE, has been proposed for connectionist temporal classification (CTC) based streaming ASR, which obtains a good balance between reduced latency and high recognition accuracy. In this paper, we present CUSIDE-T, which successfully adapts the CUSIDE method over the recurrent neural network transducer (RNN-T) ASR architecture, instead of being based on the CTC architecture. We also incorporate language model rescoring in CUSIDE-T to further enhance accuracy, while only bringing a small additional latency. Extensive experiments are conducted over the AISHELL-1, WenetSpeech and SpeechIO datasets, comparing CUSIDE-T and U2++ (both based on RNN-T). U2++ is an existing counterpart of chunk based streaming ASR method. It is shown that CUSIDE-T achieves superior accuracy performance for streaming ASR, with equal settings of latency.
Abstract:There exist three approaches for multilingual and crosslingual automatic speech recognition (MCL-ASR) - supervised pre-training with phonetic or graphemic transcription, and self-supervised pre-training. We find that pre-training with phonetic supervision has been underappreciated so far for MCL-ASR, while conceptually it is more advantageous for information sharing between different languages. This paper explores the approach of pre-training with weakly phonetic supervision towards data-efficient MCL-ASR, which is called Whistle. We relax the requirement of gold-standard human-validated phonetic transcripts, and obtain International Phonetic Alphabet (IPA) based transcription by leveraging the LanguageNet grapheme-to-phoneme (G2P) models. We construct a common experimental setup based on the CommonVoice dataset, called CV-Lang10, with 10 seen languages and 2 unseen languages. A set of experiments are conducted on CV-Lang10 to compare, as fair as possible, the three approaches under the common setup for MCL-ASR. Experiments demonstrate the advantages of phoneme-based models (Whistle) for MCL-ASR, in terms of speech recognition for seen languages, crosslingual performance for unseen languages with different amounts of few-shot data, overcoming catastrophic forgetting, and training efficiency.It is found that when training data is more limited, phoneme supervision can achieve better results compared to subword supervision and self-supervision, thereby providing higher data-efficiency. To support reproducibility and promote future research along this direction, we will release the code, models and data for the whole pipeline of Whistle at https://github.com/thu-spmi/CAT upon publication.
Abstract:Surround-view depth estimation is a crucial task aims to acquire the depth maps of the surrounding views. It has many applications in real world scenarios such as autonomous driving, AR/VR and 3D reconstruction, etc. However, given that most of the data in the autonomous driving dataset is collected in daytime scenarios, this leads to poor depth model performance in the face of out-of-distribution(OoD) data. While some works try to improve the robustness of depth model under OoD data, these methods either require additional training data or lake generalizability. In this report, we introduce the DINO-SD, a novel surround-view depth estimation model. Our DINO-SD does not need additional data and has strong robustness. Our DINO-SD get the best performance in the track4 of ICRA 2024 RoboDepth Challenge.
Abstract:In the realm of autonomous driving, robust perception under out-of-distribution conditions is paramount for the safe deployment of vehicles. Challenges such as adverse weather, sensor malfunctions, and environmental unpredictability can severely impact the performance of autonomous systems. The 2024 RoboDrive Challenge was crafted to propel the development of driving perception technologies that can withstand and adapt to these real-world variabilities. Focusing on four pivotal tasks -- BEV detection, map segmentation, semantic occupancy prediction, and multi-view depth estimation -- the competition laid down a gauntlet to innovate and enhance system resilience against typical and atypical disturbances. This year's challenge consisted of five distinct tracks and attracted 140 registered teams from 93 institutes across 11 countries, resulting in nearly one thousand submissions evaluated through our servers. The competition culminated in 15 top-performing solutions, which introduced a range of innovative approaches including advanced data augmentation, multi-sensor fusion, self-supervised learning for error correction, and new algorithmic strategies to enhance sensor robustness. These contributions significantly advanced the state of the art, particularly in handling sensor inconsistencies and environmental variability. Participants, through collaborative efforts, pushed the boundaries of current technologies, showcasing their potential in real-world scenarios. Extensive evaluations and analyses provided insights into the effectiveness of these solutions, highlighting key trends and successful strategies for improving the resilience of driving perception systems. This challenge has set a new benchmark in the field, providing a rich repository of techniques expected to guide future research in this field.
Abstract:Mesh denoising, aimed at removing noise from input meshes while preserving their feature structures, is a practical yet challenging task. Despite the remarkable progress in learning-based mesh denoising methodologies in recent years, their network designs often encounter two principal drawbacks: a dependence on single-modal geometric representations, which fall short in capturing the multifaceted attributes of meshes, and a lack of effective global feature aggregation, hindering their ability to fully understand the mesh's comprehensive structure. To tackle these issues, we propose SurfaceFormer, a pioneering Transformer-based mesh denoising framework. Our first contribution is the development of a new representation known as Local Surface Descriptor, which is crafted by establishing polar systems on each mesh face, followed by sampling points from adjacent surfaces using geodesics. The normals of these points are organized into 2D patches, mimicking images to capture local geometric intricacies, whereas the poles and vertex coordinates are consolidated into a point cloud to embody spatial information. This advancement surmounts the hurdles posed by the irregular and non-Euclidean characteristics of mesh data, facilitating a smooth integration with Transformer architecture. Next, we propose a dual-stream structure consisting of a Geometric Encoder branch and a Spatial Encoder branch, which jointly encode local geometry details and spatial information to fully explore multimodal information for mesh denoising. A subsequent Denoising Transformer module receives the multimodal information and achieves efficient global feature aggregation through self-attention operators. Our experimental evaluations demonstrate that this novel approach outperforms existing state-of-the-art methods in both objective and subjective assessments, marking a significant leap forward in mesh denoising.
Abstract:Space robots have played a critical role in autonomous maintenance and space junk removal. Multi-arm space robots can efficiently complete the target capture and base reorientation tasks due to their flexibility and the collaborative capabilities between the arms. However, the complex coupling properties arising from both the multiple arms and the free-floating base present challenges to the motion planning problems of multi-arm space robots. We observe that the octopus elegantly achieves similar goals when grabbing prey and escaping from danger. Inspired by the distributed control of octopuses' limbs, we develop a multi-level decentralized motion planning framework to manage the movement of different arms of space robots. This motion planning framework integrates naturally with the multi-agent reinforcement learning (MARL) paradigm. The results indicate that our method outperforms the previous method (centralized training). Leveraging the flexibility of the decentralized framework, we reassemble policies trained for different tasks, enabling the space robot to complete trajectory planning tasks while adjusting the base attitude without further learning. Furthermore, our experiments confirm the superior robustness of our method in the face of external disturbances, changing base masses, and even the failure of one arm.
Abstract:Sampling is widely used in various point cloud tasks as it can effectively reduce resource consumption. Recently, some methods have proposed utilizing neural networks to optimize the sampling process for various task requirements. Currently, deep downsampling methods can be categorized into two main types: generative-based and score-based. Generative-based methods directly generate sampled point clouds using networks, whereas score-based methods assess the importance of points according to specific rules and then select sampled point clouds based on their scores. However, these methods often result in noticeable clustering effects in high-intensity feature areas, compromising their ability to preserve small-scale features and leading to the loss of some structures, thereby affecting the performance of subsequent tasks. In this paper, we propose REPS, a reconstruction-based scoring strategy that evaluates the importance of each vertex by removing and reconstructing them using surrounding vertices. Our reconstruction process comprises point reconstruction and shape reconstruction. The two aforementioned reconstruction methods effectively evaluate the importance of vertices by removing them at different scales for reconstruction. These reconstructions ensure that our method maintains the overall geometric features of the point cloud and avoids disturbing small-scale structures during sampling. Additionally, we propose the Global-Local Fusion Attention (GLFA) module, which aggregates local and global attention features of point clouds, ensuring high-quality reconstruction and sampling effects. Our method outperforms previous approaches in preserving the structural features of the sampled point clouds. Furthermore, abundant experimental results demonstrate the superior performance of our method across various common tasks.
Abstract:Models of various NLP tasks have been shown to exhibit stereotypes, and the bias in the question answering (QA) models is especially harmful as the output answers might be directly consumed by the end users. There have been datasets to evaluate bias in QA models, while bias mitigation technique for the QA models is still under-explored. In this work, we propose BMBI, an approach to mitigate the bias of multiple-choice QA models. Based on the intuition that a model would lean to be more biased if it learns from a biased example, we measure the bias level of a query instance by observing its influence on another instance. If the influenced instance is more biased, we derive that the query instance is biased. We then use the bias level detected as an optimization objective to form a multi-task learning setting in addition to the original QA task. We further introduce a new bias evaluation metric to quantify bias in a comprehensive and sensitive way. We show that our method could be applied to multiple QA formulations across multiple bias categories. It can significantly reduce the bias level in all 9 bias categories in the BBQ dataset while maintaining comparable QA accuracy.
Abstract:NeurAlly-Decomposed Oracle (NADO) is a powerful approach for controllable generation with large language models. Differentiating from finetuning/prompt tuning, it has the potential to avoid catastrophic forgetting of the large base model and achieve guaranteed convergence to an entropy-maximized closed-form solution without significantly limiting the model capacity. Despite its success, several challenges arise when applying NADO to more complex scenarios. First, the best practice of using NADO for the composition of multiple control signals is under-explored. Second, vanilla NADO suffers from gradient vanishing for low-probability control signals and is highly reliant on the forward-consistency regularization. In this paper, we study the aforementioned challenges when using NADO theoretically and empirically. We show we can achieve guaranteed compositional generalization of NADO with a certain practice, and propose a novel alternative parameterization of NADO to perfectly guarantee the forward-consistency. We evaluate the improved training of NADO, i.e. NADO++, on CommonGen. Results show that NADO++ improves the effectiveness of the algorithm in multiple aspects.
Abstract:Automatic melody-to-lyric generation is a task in which song lyrics are generated to go with a given melody. It is of significant practical interest and more challenging than unconstrained lyric generation as the music imposes additional constraints onto the lyrics. The training data is limited as most songs are copyrighted, resulting in models that underfit the complicated cross-modal relationship between melody and lyrics. In this work, we propose a method for generating high-quality lyrics without training on any aligned melody-lyric data. Specifically, we design a hierarchical lyric generation framework that first generates a song outline and second the complete lyrics. The framework enables disentanglement of training (based purely on text) from inference (melody-guided text generation) to circumvent the shortage of parallel data. We leverage the segmentation and rhythm alignment between melody and lyrics to compile the given melody into decoding constraints as guidance during inference. The two-step hierarchical design also enables content control via the lyric outline, a much-desired feature for democratizing collaborative song creation. Experimental results show that our model can generate high-quality lyrics that are more on-topic, singable, intelligible, and coherent than strong baselines, for example SongMASS, a SOTA model trained on a parallel dataset, with a 24% relative overall quality improvement based on human ratings. O