Search

We introduce "Generative Fusion Decoding" (GFD), a novel shallow fusion framework, utilized to integrate Large Language Models (LLMs) into multi-modal text recognition systems such as automatic speech recognition (ASR) and optical character recognition (OCR). We derive the formulas necessary to enable GFD to operate across mismatched token spaces of different models by mapping text token space to byte token space, enabling seamless fusion during the decoding process. The framework is plug-and-play, compatible with various auto-regressive models, and does not require re-training for feature alignment, thus overcoming limitations of previous fusion techniques. We highlight three main advantages of GFD: First, by simplifying the complexity of aligning different model sample spaces, GFD allows LLMs to correct errors in tandem with the recognition model, reducing computation latencies. Second, the in-context learning ability of LLMs is fully capitalized by GFD, increasing robustness in long-form speech recognition and instruction aware speech recognition. Third, GFD enables fusing recognition models deficient in Chinese text recognition with LLMs extensively trained on Chinese. Our evaluation demonstrates that GFD significantly improves performance in ASR and OCR tasks, with ASR reaching state-of-the-art in the NTUML2021 benchmark. GFD provides a significant step forward in model integration, offering a unified solution that could be widely applicable to leveraging existing pre-trained models through step by step fusion.

Breeze-7B is an open-source language model based on Mistral-7B, designed to address the need for improved language comprehension and chatbot-oriented capabilities in Traditional Chinese. This technical report provides an overview of the additional pretraining, finetuning, and evaluation stages for the Breeze-7B model. The Breeze-7B family of base and chat models exhibits good performance on language comprehension and chatbot-oriented tasks, reaching the top in several benchmarks among models comparable in its complexity class.

Many machine learning problems lack strong convexity properties. Fortunately, recent studies have revealed that first-order algorithms also enjoy linear convergences under various weaker regularity conditions. While the relationship among different conditions for convex and smooth functions is well understood, it is not the case for the nonsmooth setting. In this paper, we go beyond convexity and smoothness, and clarify the connections among common regularity conditions (including $\textit{strong convexity, restricted secant inequality, subdifferential error bound, Polyak-{\L}ojasiewicz inequality, and quadratic growth}$) in the class of weakly convex functions. In addition, we present a simple and modular proof for the linear convergence of the $\textit{proximal point method}$ (PPM) for convex (possibly nonsmooth) optimization using these regularity conditions. The linear convergence also holds when the subproblems of PPM are solved inexactly with a proper control of inexactness., Comment: 26 pages and 2 figures

The evaluation of large language models is an essential task in the field of language understanding and generation. As language models continue to advance, the need for effective benchmarks to assess their performance has become imperative. In the context of Traditional Chinese, there is a scarcity of comprehensive and diverse benchmarks to evaluate the capabilities of language models, despite the existence of certain benchmarks such as DRCD, TTQA, CMDQA, and FGC dataset. To address this gap, we propose a novel set of benchmarks that leverage existing English datasets and are tailored to evaluate language models in Traditional Chinese. These benchmarks encompass a wide range of tasks, including contextual question-answering, summarization, classification, and table understanding. The proposed benchmarks offer a comprehensive evaluation framework, enabling the assessment of language models' capabilities across different tasks. In this paper, we evaluate the performance of GPT-3.5, Taiwan-LLaMa-v1.0, and Model 7-C, our proprietary model, on these benchmarks. The evaluation results highlight that our model, Model 7-C, achieves performance comparable to GPT-3.5 with respect to a part of the evaluated capabilities. In an effort to advance the evaluation of language models in Traditional Chinese and stimulate further research in this field, we have open-sourced our benchmark and opened the model for trial.

In this work, we propose a method to create domain-sensitive speech recognition models that utilize textual domain information by conditioning its generation on a given text prompt. This is accomplished by fine-tuning a pre-trained, end-to-end model (Whisper) to learn from demonstrations with prompt examples. We show that this ability can be generalized to different domains and even various prompt contexts, with our model gaining a Word Error Rate (WER) reduction of up to 33% on unseen datasets from various domains, such as medical conversation, air traffic control communication, and financial meetings. Considering the limited availability of audio-transcript pair data, we further extend our method to text-only fine-tuning to achieve domain sensitivity as well as domain adaptation. We demonstrate that our text-only fine-tuned model can also attend to various prompt contexts, with the model reaching the most WER reduction of 29% on the medical conversation dataset., Comment: F-T Liao and Y-C Chan contributed equally; paper accepted to ASRU2023; code and model weights available in https://github.com/mtkresearch/clairaudience

The spectral bundle method developed by Helmberg and Rendl is well-established for solving large-scale semidefinite programs (SDPs) in the dual form, especially when the SDPs admit $\textit{low-rank primal solutions}$. Under mild regularity conditions, a recent result by Ding and Grimmer has established fast linear convergence rates when the bundle method captures $\textit{the rank of primal solutions}$. In this paper, we present an overview and comparison of spectral bundle methods for solving both $\textit{primal}$ and $\textit{dual}$ SDPs. In particular, we introduce a new family of spectral bundle methods for solving SDPs in the $\textit{primal}$ form. The algorithm developments are parallel to those by Helmberg and Rendl, mirroring the elegant duality between primal and dual SDPs. The new family of spectral bundle methods also achieves linear convergence rates for primal feasibility, dual feasibility, and duality gap when the algorithm captures $\textit{the rank of the dual solutions}$. Therefore, the original spectral bundle method by Helmberg and Rendl is well-suited for SDPs with $\textit{low-rank primal solutions}$, while on the other hand, our new spectral bundle method works well for SDPs with $\textit{low-rank dual solutions}$. These theoretical findings are supported by a range of large-scale numerical experiments. Finally, we demonstrate that our new spectral bundle method achieves state-of-the-art efficiency and scalability for solving polynomial optimization compared to a set of baseline solvers $\textsf{SDPT3}$, $\textsf{MOSEK}$, $\textsf{CDCS}$, and $\textsf{SDPNAL+}$., Comment: 57 pages, 4 figures, and 4 tables

Secondary Ions Mass Spectroscopy and Hall effect measurements were performed on boron doped silicon with concentration between 0.02 at.% and 12 at.%. Ultra-high boron doping was made by saturating the chemisorption sites of a Si wafer with BCl3, followed by nanosecond laser anneal (Gas Immersion Laser Doping). The boron concentration varies thus nearly linearly with the number of process repetitions. However, it is not the case for the hole concentration which tends to saturate at high boron concentration. The difference between boron and hole concentration increases as the square of boron concentration, pointing towards the formation of boron pairs as the dominant contribution to the increase of inactive boron.

In this paper, we propose iterative inner/outer approximations based on a recent notion of block factor-width-two matrices for solving semidefinite programs (SDPs). Our inner/outer approximating algorithms generate a sequence of upper/lower bounds of increasing accuracy for the optimal SDP cost. The block partition in our algorithms offers flexibility in terms of both numerical efficiency and solution quality, which includes the approach of scaled diagonally dominance (SDD) approximation as a special case. We discuss both the theoretical results and numerical implementation in detail. Our main theorems guarantee that the proposed iterative algorithms generate monotonically decreasing upper (increasing lower) bounds. Extensive numerical results confirm our findings., Comment: 8 pages, 6 figures. Code is available through https://github.com/soc-ucsd/Iterative_SDPfw

In meta-learning, the knowledge learned from previous tasks is transferred to new ones, but this transfer only works if tasks are related. Sharing information between unrelated tasks might hurt performance, and it is unclear how to transfer knowledge across tasks with a hierarchical structure. Our research extends a model agnostic meta-learning model, MAML, by exploiting hierarchical task relationships. Our algorithm, TreeMAML, adapts the model to each task with a few gradient steps, but the adaptation follows the hierarchical tree structure: in each step, gradients are pooled across tasks clusters, and subsequent steps follow down the tree. We also implement a clustering algorithm that generates the tasks tree without previous knowledge of the task structure, allowing us to make use of implicit relationships between the tasks. We show that the new algorithm, which we term TreeMAML, performs better than MAML when the task structure is hierarchical for synthetic experiments. To study the performance of the method in real-world data, we apply this method to Natural Language Understanding, we use our algorithm to finetune Language Models taking advantage of the language phylogenetic tree. We show that TreeMAML improves the state of the art results for cross-lingual Natural Language Inference. This result is useful, since most languages in the world are under-resourced and the improvement on cross-lingual transfer allows the internationalization of NLP models. This results open the window to use this algorithm in other real-world hierarchical datasets., Comment: Updated version of paper in EACL workshop: Adapt-NLP 2021

During software front-end development, the work to convert Graphical User Interface(GUI) image to the corresponding front-end code is an inevitable tedious work. There have been some attempts to make this work to be automatic. However, the GUI code generated by these models is not accurate due to the lack of attention mechanism guidance. To solve this problem, we propose PixCoder based on an artificially supervised attention mechanism. The approach is to train a neural network to predict the style sheets in the input GUI image and then output a vector. PixCoder generate the GUI code targeting specific platform according to the output vector. The experimental results have shown the accuracy of the GUI code generated by PixCoder is over 95%.

Online media outlets adopt clickbait techniques to lure readers to click on articles in a bid to expand their reach and subsequently increase revenue through ad monetization. As the adverse effects of clickbait attract more and more attention, researchers have started to explore machine learning techniques to automatically detect clickbaits. Previous work on clickbait detection assumes that all the training data is available locally during training. In many real-world applications, however, training data is generally distributedly stored by different parties (e.g., different parties maintain data with different feature spaces), and the parties cannot share their data with each other due to data privacy issues. It is challenging to build models of high-quality federally for detecting clickbaits effectively without data sharing. In this paper, we propose a federated training framework, which is called federated hierarchical hybrid networks, to build clickbait detection models, where the titles and contents are stored by different parties, whose relationships must be exploited for clickbait detection. We empirically demonstrate that our approach is effective by comparing our approach to the state-of-the-art approaches using datasets from social media., Comment: 10 pages