An official website of the United States government
Long document re-ranking has been a challenging problem for neural re-rankers based on deep language models like BERT. Early work breaks the documents into short passage-like chunks. These chunks are independently mapped to scalar scores or latent vectors, which are then pooled into a final relevance score. These encode-and-pool methods however inevitably introduce an information bottleneck: the low dimension representations. In this paper, we propose instead to model full query-to-document interaction, leveraging the attention operation and modular Transformer re-ranker framework. First, document chunks are encoded independently with an encoder module. An interaction module then encodes the query and performs joint attention from the query to all document chunk representations. We demonstrate that the model can use this new degree of freedom to aggregate important information from the entire document. Our experiments show that this design produces effective re-ranking on two classical IR collections Robust04 and ClueWeb09, and a large-scale supervised collection MS-MARCO document ranking. CCS CONCEPTS • Information systems
more »
« less
Modularized transformer-based ranking framework
Recent innovations in Transformer-based ranking models have advanced the state-of-the-art in information retrieval. However, these Transformers are computationally expensive, and their opaque hidden states make it hard to understand the ranking process. In this work, we modularize the Transformer ranker into separate modules for text representation and interaction. We show how this design enables substantially faster ranking using offline pre-computed representations and light-weight online interactions. The modular design is also easier to interpret and sheds light on the ranking process in Transformer rankers.
more »
« less
- Award ID(s):
- 1815528
- PAR ID:
- 10273582
- Date Published:
- Journal Name:
- Conference on Empirical Methods in Natural Language Processing
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Personal assistant systems, such as Apple Siri, Google Now, Amazon Alexa, and Microsoft Cortana, are becoming ever more widely used. Understanding user intent such as clarification questions, potential answers and user feedback in information-seeking conversations is critical for retrieving good responses. In this paper, we analyze user intent patterns in information-seeking conversations and propose an intent-aware neural response ranking model ``IART'', which refers to ``Intent-Aware Ranking with Transformers''. IART is built on top of the integration of user intent modeling and language representation learning with the Transformer architecture, which relies entirely on a self-attention mechanism instead of recurrent nets. It incorporates intent-aware utterance attention to derive an importance weighting scheme of utterances in conversation context with the aim of better conversation history understanding. We conduct extensive experiments with three information-seeking conversation data sets including both standard benchmarks and commercial data. Our proposed model outperforms all baseline methods with respect to a variety of metrics. We also perform case studies and analysis of learned user intent and its impact on response ranking in information-seeking conversations to provide interpretation of results. Our research findings provide insights on intent-aware neural ranking models based on Transformers for response selection, and have implications for the design of the next generation of information-seeking conversation systems.more » « less
-
Development of the new generation of high power and high frequency power electronic switches along with the need for compact controllable converters for utilization of distributed energy resources in the grid, have led to significant developments in the area of solid state transformers in the last years. The design process of a high frequency transformer as the main element in the solid state transformer is illustrated in this article. A multi winding transformer for multiport SST application is designed, studied and built in this research. In a MPSST several windings feed the core. As the result, coupling coefficient between each pair of windings, become an important factor which is studied in this study. Since the transformer is designed for high frequency applications, the power loss in the wire and core of the transformer increases as the result of higher skin effect and eddy current loss in high frequency. Three important factors in the design of HF transformer for MPSST are discussed in the paper. First, four different possible core materials are compared based on their flux density, frequency range, loss and price. Then the cable selection is illustrated and finally, different winding placement and distribution on the same core are suggested and the inductance and coupling coefficient matrices are calculated using ANSYS Maxwell 3D simulation. The transformer is built in the lab and the inductance values matches the expected values from the simulation.more » « less
-
Abstract MotivationProteins interact to form complexes to carry out essential biological functions. Computational methods such as AlphaFold-multimer have been developed to predict the quaternary structures of protein complexes. An important yet largely unsolved challenge in protein complex structure prediction is to accurately estimate the quality of predicted protein complex structures without any knowledge of the corresponding native structures. Such estimations can then be used to select high-quality predicted complex structures to facilitate biomedical research such as protein function analysis and drug discovery. ResultsIn this work, we introduce a new gated neighborhood-modulating graph transformer to predict the quality of 3D protein complex structures. It incorporates node and edge gates within a graph transformer framework to control information flow during graph message passing. We trained, evaluated and tested the method (called DProQA) on newly-curated protein complex datasets before the 15th Critical Assessment of Techniques for Protein Structure Prediction (CASP15) and then blindly tested it in the 2022 CASP15 experiment. The method was ranked 3rd among the single-model quality assessment methods in CASP15 in terms of the ranking loss of TM-score on 36 complex targets. The rigorous internal and external experiments demonstrate that DProQA is effective in ranking protein complex structures. Availability and implementationThe source code, data, and pre-trained models are available at https://github.com/jianlin-cheng/DProQA.more » « less
-
Graph Neural Networks (GNNs) that are based on the message passing (MP) paradigm generally exchange information between 1-hop neighbors to build node representations at each layer. In principle, such networks are not able to capture long-range interactions (LRI) that may be desired or necessary for learning a given task on graphs. Recently, there has been an increasing interest in development of Transformer-based methods for graphs that can consider full node connectivity beyond the original sparse structure, thus enabling the modeling of LRI. However, MP-GNNs that simply rely on 1-hop message passing often fare better in several existing graph benchmarks when combined with positional feature representations, among other innovations, hence limiting the perceived utility and ranking of Transformer-like architectures. Here, we present the Long Range Graph Benchmark (LRGB)1 with 5 graph learning datasets: PascalVOC-SP, COCO-SP, PCQM-Contact, Peptides-func and Peptides-struct that arguably require LRI reasoning to achieve strong performance in a given task. We benchmark both baseline GNNs and Graph Transformer networks to verify that the models which capture long-range dependencies perform significantly better on these tasks. Therefore, these datasets are suitable for benchmarking and exploration of MP-GNNs and Graph Transformer architectures that are intended to capture LRI.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
