爱可可AI前沿推介(1.18)

LG - 机器学习 CV - 计算机视觉 CL - 计算与语言 IR - 信息检索

1、[CL] Streaming Punctuation: A Novel Punctuation Technique Leveraging Bidirectional Context for Continuous Speech Recognition
2、[IR] Taking Search to Task
3、[CL] Improving Generalization of Adapter-Based Cross-lingual Transfer with Scheduled Unfreezing
4、[LG] A survey and taxonomy of loss functions in machine learning
5、[IR] Causal Inference for Recommendation: Foundations, Methods and Applications

摘要：连续语音识别基于双向上下文的标点标记、关于搜索中任务的综述、通过规划解冻改善基于适配器的跨语言迁移的泛化、机器学习损失函数综述、面向推荐的因果推断综述

1、[CL] Streaming Punctuation: A Novel Punctuation Technique Leveraging Bidirectional Context for Continuous Speech Recognition

P Behre, S Tan, P Varadharajan, S Chang
[Microsoft]

流标点：连续语音识别基于双向上下文的标点标记

要点:

1. 提出一种用动态解码窗口对 ASR 输出标点标记或重标记的流方法，以提高跨场景的标点打标和分割精度；
2. 基于动态解码窗口对 ASR 输出进行流式的标点标记或重标记；
3. 解决了过度分割问题，改善了机器翻译下游任务结果，对各模型架构具有鲁棒性。

一句话总结:
提出的流方法提高了 ASR 标点和分割精度，减少了过度分割，改善了机器翻译结果，对各模型架构具有鲁棒性。

摘要：
虽然英语语音识别单词错误率(WER)已经达到人类水平，但由于不规则的暂停模式或说话速度缓慢，语音输入和会议转录等连续语音识别场景仍然受到分割和标点符号问题的影响。Transformer 序列标记模型在捕获长双向上下文方面是有效的，这对自动标点标记至关重要。然而，自动语音识别(ASR)生产系统受到实时需求的限制，因此在做出标点决策时很难包含正确的上下文。ASR解码器生成的片段中的上下文可能有所帮助，但会限制连续语音会话的整体标点标记性能。本文提出一种用动态解码窗口对ASR输出进行标点标记或重标记的流方法，并测量其对跨场景标点和分割精度的影响。新系统解决了过度分割问题，将分割F0.5得分提高了13.9%。流标点在机器翻译(MT)下游任务中实现了0.66的平均BLEU得分提升。

While speech recognition Word Error Rate (WER) has reached human parity for English, continuous speech recognition scenarios such as voice typing and meeting transcriptions still suffer from segmentation and punctuation problems, resulting from irregular pausing patterns or slow speakers. Transformer sequence tagging models are effective at capturing long bi-directional context, which is crucial for automatic punctuation. Automatic Speech Recognition (ASR) production systems, however, are constrained by real-time requirements, making it hard to incorporate the right context when making punctuation decisions. Context within the segments produced by ASR decoders can be helpful but limiting in overall punctuation performance for a continuous speech session. In this paper, we propose a streaming approach for punctuation or re-punctuation of ASR output using dynamic decoding windows and measure its impact on punctuation and segmentation accuracy across scenarios. The new system tackles over-segmentation issues, improving segmentation F0.5-score by 13.9%. Streaming punctuation achieves an average BLEUscore improvement of 0.66 for the downstream task of Machine Translation (MT).

https://arxiv.org/abs/2301.03819

2、[IR] Taking Search to Task

C Shah, R W. White, P Thomas, B Mitra, S Sarkar, N Belkin
[University of Washington & Microsoft Research & Rutgers University]

关于搜索中任务的综述

要点:

1. 提供了各种观点，说明 IR 中任务的最新进展，衍生和使用任务信息的瓶颈，以及探索方向；
2. 综合了理解、提取和解决以任务为重点的搜索的历史和当前观点；
3. 提出一种新的框架性方法，用类似树状的结构，以实现不同的解释和应用。

一句话总结:
本文阐述了IR中任务的最新进展，衍生和使用任务信息的一些瓶颈，提出一种采用树状结构的任务框架性方法，以帮助重振基于任务的IR的兴趣和未来工作，特别是面向对话智能体和主动IR等新兴领域。

摘要：
长期以来，信息检索(IR)中任务的重要性一直被讨论，以不同的方式处理，经常被忽视，也经常被重新审视。几十年来，学者们证明了用户任务在用户如何以及为什么参与搜索以及搜索系统应该做些什么来提供帮助方面所起的作用。但在大多数情况下，IR社区过于专注于查询处理和假设搜索任务是用户查询的集合，往往忽视这种假设是否或如何帮用户解决任务。随着对话智能体和主动式IR等新领域的出现，理解和解决用户任务比以往任何时候都更加重要。本文提供了各种观点，说明IR中任务的最新进展，衍生和使用任务信息的一些瓶颈是什么，以及如何从这里继续前进。除了涵盖相关文献外，本文还综合了理解、提取和解决以任务为重点的搜索的历史和当前观点。为了奠定该领域正在进行的和未来研究的基础，本文提出了一种新的框架性方法，使用类似树状的结构，以实现不同的解释和应用。本文结合了想法和过去的工作、未来研究的建议以及对技术、社会和伦理考虑的看法，旨在帮助重振基于任务的IR的兴趣和未来工作。

The importance of tasks in information retrieval (IR) has been long argued for, addressed in different ways, often ignored, and frequently revisited. For decades, scholars made a case for the role that a user's task plays in how and why that user engages in search and what a search system should do to assist. But for the most part, the IR community has been too focused on query processing and assuming a search task to be a collection of user queries, often ignoring if or how such an assumption addresses the users accomplishing their tasks. With emerging areas of conversational agents and proactive IR, understanding and addressing users' tasks has become more important than ever before. In this paper, we provide various perspectives on where the state-of-the-art is with regard to tasks in IR, what are some of the bottlenecks in deriving and using task information, and how do we go forward from here. In addition to covering relevant literature, the paper provides a synthesis of historical and current perspectives on understanding, extracting, and addressing task-focused search. To ground ongoing and future research in this area, we present a new framing device for tasks using a tree-like structure and various moves on that structure that allow different interpretations and applications. Presented as a combination of synthesis of ideas and past works, proposals for future research, and our perspectives on technical, social, and ethical considerations, this paper is meant to help revitalize the interest and future work in task-based IR.

https://arxiv.org/abs/2301.05046

3、[CL] Improving Generalization of Adapter-Based Cross-lingual Transfer with Scheduled Unfreezing

C C Liu, J Pfeiffer...
[Technical University of Darmstadt & Google Research & University of Cambridge]

通过规划解冻改善基于适配器的跨语言迁移的泛化

要点:

1. 证明在跨语言迁移的任务适应器训练中计划解冻将性能差距缩小到完整的模型微调；
2. 提出包含几种现有方法的通用规划解冻框架；
3. 提出一种基于 Fisher Information 的规划解冻方法(FUN)，其性能与启发式方法相当。

一句话总结:
旨在用规划解冻算法改进基于适配器的跨语言迁移的泛化，证明可缩小其与完全微调间的差距，同时仍然实现最先进的迁移性能。

摘要：
标准的语言模型微调通常在分布内数据上表现良好，但分布漂移时泛化会受到影响。本文目标是在这种发生跨语言分布漂移的情况下，改进基于适配器的跨语言任务迁移的泛化。本文研究了用于在跨语言迁移中微调任务适配器的规划解冻算法，该算法最初旨在减少迁移学习中的灾难性遗忘。实验表明，规划解冻方法缩小了其与完全微调的差距，并实现了最先进的迁移性能，表明这些方法不仅仅是减轻灾难性遗忘。接下来，为了更深入地研究这些实证发现，本文研究了用Fisher Information规划解冻的学习动态。深入实验表明，与标准微调相比，规划解冻会导致不同的学习动态，并证明Fisher Information在训练期间的动态与跨语言泛化性能相关。本文还提出了一种通用的规划解冻算法，与标准微调相比，该算法在四个数据集上平均提高了2个百分点，并为启发式解冻规划(即启发式时间表隐式地最大化Fisher信息)提供了强有力的经验证据。

Standard fine-tuning of language models typically performs well on in-distribution data, but suffers with generalization to distribution shifts. In this work, we aim to improve generalization of adapter-based cross-lingual task transfer where such cross-language distribution shifts are imminent. We investigate scheduled unfreezing algorithms -- originally proposed to mitigate catastrophic forgetting in transfer learning -- for fine-tuning task adapters in cross-lingual transfer. Our experiments show that scheduled unfreezing methods close the gap to full fine-tuning and achieve state-of-the-art transfer performance, suggesting that these methods can go beyond just mitigating catastrophic forgetting. Next, aiming to delve deeper into those empirical findings, we investigate the learning dynamics of scheduled unfreezing using Fisher Information. Our in-depth experiments reveal that scheduled unfreezing induces different learning dynamics compared to standard fine-tuning, and provide evidence that the dynamics of Fisher Information during training correlate with cross-lingual generalization performance. We additionally propose a general scheduled unfreezing algorithm that achieves an average of 2 points improvement over four datasets compared to standard fine-tuning and provides strong empirical evidence for a theory-based justification of the heuristic unfreezing schedule (i.e., the heuristic schedule is implicitly maximizing Fisher Information). Our code will be publicly available.

https://arxiv.org/abs/2301.05487

4、[LG] A survey and taxonomy of loss functions in machine learning

L Ciampiconi, A Elwood, M Leonardi, A Mohamed, A Rozza
[lastminute.com group]

机器学习损失函数综述

要点:

1. 对各种机器学习应用的33种常用损失函数进行调研，包括分类、回归、排序、样本生成和基于能源建模；
2. 损失函数的直观分类，按任务、学习范式和基本策略来进行组织；
3. 为初学者和高级机器学习从业者在为他们的问题定义适当损失函数时提供使用参考。

一句话总结:
对各种机器学习应用的33种常用损失函数进行调研，按易于理解的分类进行整理，作为从业者在为问题定义适当损失函数时提供参考。

摘要：
大多数最先进的机器学习技术，都围绕着损失函数的优化。因此，定义适当的损失函数对于成功解决该领域的问题至关重要。本文对各种不同应用中最常用的损失函数进行了调研，分为分类、回归、排序、样本生成和基于能源建模。本文将33种不同的损失函数，组织成容易理解的分类。每种损失函数都有其理论支持，本文描述了其最适合使用的场景。本综述旨在为初学者和高级机器学习从业者提供最基本的损失函数参考。

Most state-of-the-art machine learning techniques revolve around the optimisation of loss functions. Defining appropriate loss functions is therefore critical to successfully solving problems in this field. We present a survey of the most commonly used loss functions for a wide range of different applications, divided into classification, regression, ranking, sample generation and energy based modelling. Overall, we introduce 33 different loss functions and we organise them into an intuitive taxonomy. Each loss function is given a theoretical backing and we describe where it is best used. This survey aims to provide a reference of the most essential loss functions for both beginner and advanced machine learning practitioners.

https://arxiv.org/abs/2301.05579

5、[IR] Causal Inference for Recommendation: Foundations, Methods and Applications

S Xu, J Ji, Y Li, Y Ge, J Tan, Y Zhang
[Rutgers University]

面向推荐的因果推断：基础、方法和应用

要点:

1. 全面回顾推荐系统中因果推理的相关文献，包括推荐系统和因果推断的基本概念及其关系；
2. 讨论完全依赖推荐系统相关性可能产生的实际问题，如公平性、可解释性、鲁棒性、偏差、回声室和可控性问题；
3. 探索推荐系统不同问题的因果方法的现有工作，包括可解释性、公平性、鲁棒性、基于提升和推荐中的公正性。

一句话总结:
全面调研面向推荐系统的因果推断的相关文献，包括基本概念、实际问题、关于各种问题的因果方法的现有工作，以及开放式问题和未来方向。

摘要：
推荐系统是各种个性化服务的重要而强大的工具。传统上，这些系统使用数据挖掘和机器学习技术根据数据中的相关性提出建议。然而，仅仅依靠相关性而不考虑潜在的因果机制可能会导致各种实际问题，如公平性、可解释性、鲁棒性、偏差、回声室和可控性问题。因此，相关领域的研究人员已经开始将因果纳入推荐系统，以解决这些问题。本综述回顾了关于推荐系统因果推断的现有文献。讨论了推荐系统和因果推断的基本概念及其关系，回顾了推荐系统中不同问题的因果方法的现有工作。最后，讨论了建议因果推断领域的未决问题和未来方向。

Recommender systems are important and powerful tools for various personalized services. Traditionally, these systems use data mining and machine learning techniques to make recommendations based on correlations found in the data. However, relying solely on correlation without considering the underlying causal mechanism may lead to various practical issues such as fairness, explainability, robustness, bias, echo chamber and controllability problems. Therefore, researchers in related area have begun incorporating causality into recommendation systems to address these issues. In this survey, we review the existing literature on causal inference in recommender systems. We discuss the fundamental concepts of both recommender systems and causal inference as well as their relationship, and review the existing work on causal methods for different problems in recommender systems. Finally, we discuss open problems and future directions in the field of causal inference for recommendations.

https://arxiv.org/abs/2301.04016