Dense passage retrieval enhances Information Retrieval (IR) by encoding queries and passages into representation space. However, passage representations often fail to be referenced by their gold queries under domain shifts, revealing a weakness in representation space. One desirable concept for representations is ”argmaxable”. Being argmaxable ensures that no representations are theoretically excluded from selection due to geometric constraints. To be argmaxable, a notable approach is to increase isotropy, where representations are evenly spread out in all directions. These findings, while desirable also for IR, focus on passage representation and not on query, making it challenging to directly apply their findings to IR. In contrast, we introduce a novel query-focused concept of ”referentiable” tailored for IR tasks, which ensures that passage representations are referenced by their gold queries. Building on this, we propose Learning Referentiable Representation (LRR), and two strategic metrics, Self-P and Self-Q, quantifying how the representations are referentiable. Our experiments compare three dense model versions: Naive, Isotropic, and Referentiable, demonstrating that LRR leads to enhanced zero-shot performance, surpassing existing naive and isotropic versions.

Advancements in dense retrieval models have brought ColBERT to prominence in Information Retrieval (IR) with its advanced interaction techniques.However, ColBERT is reported to frequently underperform in zero-shot scenarios, where traditional techniques such as BM25 still exceed it.Addressing this, we propose to balance representation isotropy and anisotropy for zero-shot model performance, based on our observations that isotropy can enhance cosine similarity computations and anisotropy may aid in generalizing to unseen data.Striking a balance between these isotropic and anisotropic qualities stands as a critical objective to refine model efficacy.Based on this, we present ours, a Hybrid Isotropy Learning (HIL) architecture that integrates isotropic and anisotropic representations.Our experiments with the BEIR benchmark show that our model significantly outperforms the baseline ColBERT model, highlighting the importance of harmonized isotropy in improving zero-shot retrieval performance.

While pre-trained language models (PLMs) have become a de-facto standard promoting the accuracy of text classification tasks, recent studies find that PLMs often predict over-confidently. Although calibration methods have been proposed, such as ensemble learning and data augmentation, most of the methods have been verified in computer vision benchmarks rather than in PLM-based text classification tasks. In this paper, we present an empirical study on confidence calibration for PLMs, addressing three categories, including confidence penalty losses, data augmentations, and ensemble methods. We find that the ensemble model overfitted to the training set shows sub-par calibration performance and also observe that PLMs trained with confidence penalty loss have a trade-off between calibration and accuracy. Building on these observations, we propose the Calibrated PLM (CALL), a combination of calibration techniques. The CALL complements shortcomings that may occur when utilizing a calibration method individually and boosts both classification and calibration accuracy. Design choices in CALL’s training procedures are extensively studied, and we provide a detailed analysis of how calibration techniques affect the calibration performance of PLMs.

For real-world language applications, detecting an out-of-distribution (OOD) sample is helpful to alert users or reject such unreliable samples. However, modern over-parameterized language models often produce overconfident predictions for both in-distribution (ID) and OOD samples. In particular, language models suffer from OOD samples with a similar semantic representation to ID samples since these OOD samples lie near the ID manifold.A rejection network can be trained with ID and diverse outlier samples to detect test OOD samples, but explicitly collecting auxiliary OOD datasets brings an additional burden for data collection. In this paper, we propose a simple but effective method called Pseudo Outlier Exposure (POE) that constructs a surrogate OOD dataset by sequentially masking tokens related to ID classes. The surrogate OOD sample introduced by POE shows a similar representation to ID data, which is most effective in training a rejection network. Our method does not require any external OOD data and can be easily implemented within off-the-shelf Transformers.A comprehensive comparison with state-of-the-art algorithms demonstrates POE’s competitiveness on several text classification benchmarks.

The generative retrieval model depends solely on the information encoded in its model parameters without external memory, its information capacity is limited and fixed. To overcome the limitation, we propose Nonparametric Decoding (Np Decoding) which can be applied to existing generative retrieval models. Np Decoding uses nonparametric contextualized vocab embeddings (external memory) rather than vanilla vocab embeddings as decoder vocab embeddings. By leveraging the contextualized vocab embeddings, the generative retrieval model is able to utilize both the parametric and nonparametric space. Evaluation over 9 datasets (8 single-hop and 1 multi-hop) in the document retrieval task shows that applying Np Decoding to generative retrieval models significantly improves the performance. We also show that Np Decoding is data- and parameter-efficient, and shows high performance in the zero-shot setting.