Recent pre-trained language models (PLMs) achieved great success on many natural language processing tasks through learning linguistic features and contextualized sentence representation. Since attributes captured in stacked layers of PLMs are not clearly identified, straightforward approaches such as embedding the last layer are commonly preferred to derive sentence representations from PLMs. This paper introduces the attention-based pooling strategy, which enables the model to preserve layer-wise signals captured in each layer and learn digested linguistic features for downstream tasks. The contrastive learning objective can adapt the layer-wise attention pooling to both unsupervised and supervised manners. It results in regularizing the anisotropic space of pre-trained embeddings and being more uniform. We evaluate our model on standard semantic textual similarity (STS) and semantic search tasks. As a result, our method improved the performance of the base contrastive learned BERT_base and variants.
JudgeMoE: Distributional Aggregation for LLM-as-a-Judge
The University of Manchester（曼彻斯特大学）；The University of Sheffield（谢菲尔德大学）；University of Pittsburgh（匹兹堡大学）；Shanghai University of Finance and Economics（上海财经大学）
Don't Judge a Book by its Cover: Testing LLMs' Robustness Under Logical Obfuscation
Manipal University Jaipur（马纳普大学贾普尔分校）；Indian Institute of Technology, Patna（印度理工学院帕坦分校）；Indian Institute of Science Education and Research, Kolkata（印度科学教育与研究学院科钦分校）
LLM-as-a-qualitative-judge: automating error analysis in natural language generation
Naver Labs Europe（纳维尔实验室欧洲分公司）；Insight Research Ireland Centre for Data Analytics, Data Science Institute, University of Galway（爱尔兰洞察研究数据分析中心、数据科学研究所、Galway大学）；University of Amsterdam（阿姆斯特丹大学）；University of Copenhagen（哥本哈根大学）；STMicroelectronics（STMicroelectronics公司）；Télécom Paris（巴黎电信学院）；Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI)（德国人工智能研究中心有限公司）；Technische Universität Berlin（柏林技术大学）；Univ. Grenoble Alpes, CNRS, Grenoble INP, LIG（格勒诺布尔阿尔卑斯大学、CNRS、格勒诺布尔INP、LIG）
Don't Judge Me by My Face : An Indirect Adversarial Approach to Remove Sensitive Information From Multimodal Neural Representation in Asynchronous Job Video Interviews
PIGLeT: Language Grounding Through Neuro-Symbolic Interaction in a 3D World
Paul G. Allen School of Computer Science & Engineering, University of Washington（华盛顿大学保罗·G·艾伦计算机科学与工程学院）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）
