Supply chain operations traditionally involve a variety of complex decision making problems. Over the last few decades, supply chains greatly benefited from advances in computation, which allowed the transition from manual processing to automation and cost-effective optimization. Nonetheless, business operators still need to spend substantial efforts in explaining and interpreting the optimization outcomes to stakeholders. Motivated by the recent advances in Large Language Models (LLMs), we study how this disruptive technology can help bridge the gap between supply chain automation and human comprehension and trust thereof. We design OptiGuide -- a framework that accepts as input queries in plain text, and outputs insights about the underlying optimization outcomes. Our framework does not forgo the state-of-the-art combinatorial optimization technology, but rather leverages it to quantitatively answer what-if scenarios (e.g., how would the cost change if we used supplier B instead of supplier A for a given demand?). Importantly, our design does not require sending proprietary data over to LLMs, which can be a privacy concern in some circumstances. We demonstrate the effectiveness of our framework on a real server placement scenario within Microsoft's cloud supply chain. Along the way, we develop a general evaluation benchmark, which can be used to evaluate the accuracy of the LLM output in other scenarios.
The state of the art in human computer conversation leaves something to be desired and, indeed, talking to a computer can be down-right annoying. This paper describes an approach to identifying ``opportunities for improvement'' in these systems by looking for abuse in the form of swear words. The premise is that humans swear at computers as a sanction and, as such, swear words represent a point of failure where the system did not behave as it should. Having identified where things went wrong, we can work backward through the transcripts and, using conversation analysis (CA) work out how things went wrong. Conversation analysis is a qualitative methodology and can appear quite alien - indeed unscientific - to those of us from a quantitative background. The paper starts with a description of Conversation analysis in its modern form, and then goes on to apply the methodology to transcripts of frustrated and annoyed users in the DARPA Communicator project. The conclusion is that there is at least one species of failure caused by the inability of the Communicator systems to handle mixed initiative at the discourse structure level. Along the way, I hope to demonstrate that there is an alternative future for computational linguistics that does not rely on larger and larger text corpora.
Cross-modal alignment plays a crucial role in vision-language pre-training (VLP) models, enabling them to capture meaningful associations across different modalities. For this purpose, inspired by the success of masked language modeling (MLM) tasks in the NLP pre-training area, numerous masked modeling tasks have been proposed for VLP to further promote cross-modal interactions. The core idea of previous masked modeling tasks is to focus on reconstructing the masked tokens based on visible context for learning local-local alignment, i.e., associations between image patches and text tokens. However, most of them pay little attention to the global semantic features generated for the masked data, resulting in a limited cross-modal alignment ability of global representations to local features of the other modality. Therefore, in this paper, we propose a novel Global and Local Semantic Completion Learning (GLSCL) task to facilitate global-local alignment and local-local alignment simultaneously. Specifically, the GLSCL task complements the missing semantics of masked data and recovers global and local features by cross-modal interactions. Our GLSCL consists of masked global semantic completion (MGSC) and masked local token completion (MLTC). MGSC promotes learning more representative global features which have a great impact on the performance of downstream tasks, and MLTC can further enhance accurate comprehension on multimodal data. Moreover, we present a flexible vision encoder, enabling our model to simultaneously perform image-text and video-text multimodal tasks. Experimental results show that our proposed method obtains state-of-the-art performance on various vision-language benchmarks, such as visual question answering, image-text retrieval, and video-text retrieval.
Open-sourced large language models (LLMs) have demonstrated remarkable efficacy in various tasks with instruction tuning. However, these models can sometimes struggle with tasks that require more specialized knowledge such as translation. One possible reason for such deficiency is that instruction tuning aims to generate fluent and coherent text that continues from a given instruction without being constrained by any task-specific requirements. Moreover, it can be more challenging for tuning smaller LLMs with lower-quality training data. To address this issue, we propose a novel framework using examples in comparison to teach LLMs to learn translation. Our approach involves presenting the model with examples of correct and incorrect translations and using a preference loss to guide the model's learning. We evaluate our method on WMT2022 test sets and show that it outperforms existing methods. Our findings offer a new perspective on fine-tuning LLMs for translation tasks and provide a promising solution for generating high-quality translations. Please refer to Github for more details: https://github.com/lemon0830/TIM.
We study parameterisation-independent closed planar curve matching as a Bayesian inverse problem. The motion of the curve is modelled via a curve on the diffeomorphism group acting on the ambient space, leading to a large deformation diffeomorphic metric mapping (LDDMM) functional penalising the kinetic energy of the deformation. We solve Hamilton's equations for the curve matching problem using the Wu-Xu element [S. Wu, J. Xu, Nonconforming finite element spaces for $2m^\text{th}$ order partial differential equations on $\mathbb{R}^n$ simplicial grids when $m=n+1$, Mathematics of Computation 88 (316) (2019) 531-551] which provides mesh-independent Lipschitz constants for the forward motion of the curve, and solve the inverse problem for the momentum using Bayesian inversion. Since this element is not affine-equivalent we provide a pullback theory which expedites the implementation and efficiency of the forward map. We adopt ensemble Kalman inversion using a negative Sobolev norm mismatch penalty to measure the discrepancy between the target and the ensemble mean shape. We provide several numerical examples to validate the approach.
Prior works about text-to-image synthesis typically concatenated the sentence embedding with the noise vector, while the sentence embedding and the noise vector are two different factors, which control the different aspects of the generation. Simply concatenating them will entangle the latent factors and encumber the generative model. In this paper, we attempt to decompose these two factors and propose Factor Decomposed Generative Adversarial Networks~(FDGAN). To achieve this, we firstly generate images from the noise vector and then apply the sentence embedding in the normalization layer for both generator and discriminators. We also design an additive norm layer to align and fuse the text-image features. The experimental results show that decomposing the noise and the sentence embedding can disentangle latent factors in text-to-image synthesis, and make the generative model more efficient. Compared with the baseline, FDGAN can achieve better performance, while fewer parameters are used.
Recent voice assistants are usually based on the cascade spoken language understanding (SLU) solution, which consists of an automatic speech recognition (ASR) engine and a natural language understanding (NLU) system. Because such approach relies on the ASR output, it often suffers from the so-called ASR error propagation. In this work, we investigate impacts of this ASR error propagation on state-of-the-art NLU systems based on pre-trained language models (PLM), such as BERT and RoBERTa. Moreover, a multimodal language understanding (MLU) module is proposed to mitigate SLU performance degradation caused by errors present in the ASR transcript. The MLU benefits from self-supervised features learned from both audio and text modalities, specifically Wav2Vec for speech and Bert/RoBERTa for language. Our MLU combines an encoder network to embed the audio signal and a text encoder to process text transcripts followed by a late fusion layer to fuse audio and text logits. We found that the proposed MLU showed to be robust towards poor quality ASR transcripts, while the performance of BERT and RoBERTa are severely compromised. Our model is evaluated on five tasks from three SLU datasets and robustness is tested using ASR transcripts from three ASR engines. Results show that the proposed approach effectively mitigates the ASR error propagation problem, surpassing the PLM models' performance across all datasets for the academic ASR engine.
Standard methods for multi-label text classification largely rely on encoder-only pre-trained language models, whereas encoder-decoder models have proven more effective in other classification tasks. In this study, we compare four methods for multi-label classification, two based on an encoder only, and two based on an encoder-decoder. We carry out experiments on four datasets -- two in the legal domain and two in the biomedical domain, each with two levels of label granularity -- and always depart from the same pre-trained model, T5. Our results show that encoder-decoder methods outperform encoder-only methods, with a growing advantage on more complex datasets and labeling schemes of finer granularity. Using encoder-decoder models in a non-autoregressive fashion, in particular, yields the best performance overall, so we further study this approach through ablations to better understand its strengths.
Decision-makers in the humanitarian sector rely on timely and exact information during crisis events. Knowing how many civilians were injured during an earthquake is vital to allocate aids properly. Information about such victim counts is often only available within full-text event descriptions from newspapers and other reports. Extracting numbers from text is challenging: numbers have different formats and may require numeric reasoning. This renders purely string matching-based approaches insufficient. As a consequence, fine-grained counts of injured, displaced, or abused victims beyond fatalities are often not extracted and remain unseen. We cast victim count extraction as a question answering (QA) task with a regression or classification objective. We compare regex, dependency parsing, semantic role labeling-based approaches, and advanced text-to-text models. Beyond model accuracy, we analyze extraction reliability and robustness which are key for this sensitive task. In particular, we discuss model calibration and investigate few-shot and out-of-distribution performance. Ultimately, we make a comprehensive recommendation on which model to select for different desiderata and data domains. Our work is among the first to apply numeracy-focused large language models in a real-world use case with a positive impact.
Current perception models for different tasks usually exist in modular forms on Unmanned Surface Vehicles (USVs), which infer extremely slowly in parallel on edge devices, causing the asynchrony between perception results and USV position, and leading to error decisions of autonomous navigation. Compared with Unmanned Ground Vehicles (UGVs), the robust perception of USVs develops relatively slowly. Moreover, most current multi-task perception models are huge in parameters, slow in inference and not scalable. Oriented on this, we propose Achelous, a low-cost and fast unified panoptic perception framework for water-surface perception based on the fusion of a monocular camera and 4D mmWave radar. Achelous can simultaneously perform five tasks, detection and segmentation of visual targets, drivable-area segmentation, waterline segmentation and radar point cloud segmentation. Besides, models in Achelous family, with less than around 5 million parameters, achieve about 18 FPS on an NVIDIA Jetson AGX Xavier, 11 FPS faster than HybridNets, and exceed YOLOX-Tiny and Segformer-B0 on our collected dataset about 5 mAP$_{\text{50-95}}$ and 0.7 mIoU, especially under situations of adverse weather, dark environments and camera failure. To our knowledge, Achelous is the first comprehensive panoptic perception framework combining vision-level and point-cloud-level tasks for water-surface perception. To promote the development of the intelligent transportation community, we release our codes in \url{https://github.com/GuanRunwei/Achelous}.