Large language models (LLMs) are now rapidly advancing and surpassing human
abilities on many natural language tasks. However, aligning these super-human
LLMs with human knowledge remains challenging because the supervision signals
from human annotators may be wrong. This issue, known as the "super-alignment"
problem, requires enhancing weak-to-strong generalization, where a strong LLM
must generalize from imperfect supervision provided by a weaker source. To
address this issue, we propose an approach to improve weak-to-strong
generalization by involving the reliability of weak supervision signals in the
alignment process. In our method, we query the weak supervisor for multiple
answers, estimate the answer reliability, and enhance the alignment process by
filtering out uncertain data or re-weighting reliable data. Experiments on four
datasets demonstrate that our methods effectively identify the quality of weak
labels and significantly enhance weak-to-strong generalization. Our work
presents effective techniques for error-robust model alignment, reducing error
propagation from noisy supervision and enhancing the accuracy and reliability
of LLMs. Codes are publicly available at
this http URL

通过提高弱监督信号的可靠性，我们的方法有效地识别了弱标签的质量，并显著提高了弱到强泛化能力，降低了噪声监督的误差传播，增强了大型语言模型的准确性和可靠性。

提升弱到强泛化性能的可靠性感知对齐

Improving Weak-to-Strong Generalization with Reliability-Aware Alignment

As the most fundamental tasks of computer vision, object detection and
segmentation have made tremendous progress in the deep learning era. Due to the
expensive manual labeling, the annotated categories in existing datasets are
often small-scale and pre-defined, i.e., state-of-the-art detectors and
segmentors fail to generalize beyond the closed-vocabulary. To resolve this
limitation, the last few years have witnessed increasing attention toward
Open-Vocabulary Detection (OVD) and Segmentation (OVS). In this survey, we
provide a comprehensive review on the past and recent development of OVD and
OVS. To this end, we develop a taxonomy according to the type of task and
methodology. We find that the permission and usage of weak supervision signals
can well discriminate different methodologies, including: visual-semantic space
mapping, novel visual feature synthesis, region-aware training,
pseudo-labeling, knowledge distillation-based, and transfer learning-based. The
proposed taxonomy is universal across different tasks, covering object
detection, semantic/instance/panoptic segmentation, 3D scene and video
understanding. In each category, its main principles, key challenges,
development routes, strengths, and weaknesses are thoroughly discussed. In
addition, we benchmark each task along with the vital components of each
method. Finally, several promising directions are provided to stimulate future
research.

计算机视觉中的目标检测和分割作为最基本的任务，在深度学习时代取得了巨大的进展。然而，由于昂贵的手动标注，现有数据集中的标注类别往往规模较小且预定义，即最先进的检测器和分割器无法推广到开放词汇之外。因此，近年来越来越多的关注集中在开放词汇检测（OVD）和分割（OVS）上。在本调研中，我们提供了对过去和最新 OVD 和 OVS 发展的全面审查。为此，我们根据任务类型和方法学开发了一个分类法。我们发现，对于不同的方法学，包括：视觉 - 语义空间映射、新颖的视觉特征合成、区域感知训练、伪标签、基于知识蒸馏和基于迁移学习的方法，弱监督信号的许可和使用可以很好地区分。所提出的分类法在不同任务之间是通用的，包括目标检测、语义 / 实例 / 全景分割、三维场景和视频理解。在每个类别中，详细讨论了其主要原则、关键挑战、发展路线、优点和缺点。此外，我们还对每个方法的关键组成部分进行了基准测试。最后，我们提供了一些有前途的方向，以激发未来的研究。

开放词汇检测和分割综述：过去、现在和未来

A Survey on Open-Vocabulary Detection and Segmentation: Past, Present,  and Future

Humans can infer the affordance of objects by extracting related contextual
preconditions for each scenario. For example, upon seeing an image of a broken
cup, we can infer that this precondition prevents the cup from being used for
drinking. Reasoning with preconditions of commonsense is studied in NLP where
the model explicitly gets the contextual precondition. However, it is unclear
if SOTA visual language models (VLMs) can extract such preconditions and infer
the affordance of objects with them. In this work, we introduce the task of
preconditioned visual language inference and rationalization (PVLIR). We
propose a learning resource based on three strategies to retrieve weak
supervision signals for the task and develop a human-verified test set for
evaluation. Our results reveal the shortcomings of SOTA VLM models in the task
and draw a road map to address the challenges ahead in improving them.

本文提出了预先条件的视觉语言推理与理性化（PVLIR）的任务，研究表明 SOTA 的 VLM 模型无法提取出这种先决条件，提出了三种策略来提取弱监督信号，并开发了经过人工验证的测试集进行评估。