arXiv subjects

Jimmy Lin

Publications and source records attributed to Jimmy Lin.

At least 199 records · Page 11Linked to original sources

arXiv2019

Exploiting Token and Path-based Representations of Code for Identifying Security-Relevant Commits

Public vulnerability databases such as CVE and NVD account for only 60% of security vulnerabilities present in open-source projects, and are known to suffer from inconsistent quality. Over the last two years, there has been considerable growth in the number of known vulnerabilities across projects available in various repositories such as NPM and Maven Central. Such an increasing risk calls for a mechanism to infer the presence of security threats in a timely manner. We propose novel hierarchical deep learning models for the identification of security-relevant commits from either the commit diff or the source code for the Java classes. By comparing the performance of our model against code2vec, a state-of-the-art model that learns from path-based representations of code, and a logistic regression baseline, we show that deep learning models show promising results in identifying security-related commits. We also conduct a comparative analysis of how various deep learning models learn across different input representations and the effect of regularization on the generalization of our models.

cs.SE↗

arXiv2019

Cross-Lingual Relevance Transfer for Document Retrieval

Recent work has shown the surprising ability of multi-lingual BERT to serve as a zero-shot cross-lingual transfer model for a number of language processing tasks. We combine this finding with a similarly-recently proposal on sentence-level relevance modeling for document retrieval to demonstrate the ability of multi-lingual BERT to transfer models of relevance across languages. Experiments on test collections in five different languages from diverse language families (Chinese, Arabic, French, Hindi, and Bengali) show that models trained with English data improve ranking quality, without any special processing, both for (non-English) mono-lingual retrieval as well as cross-lingual retrieval.

cs.IR↗

arXiv2019

What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning

Pretrained transformer-based language models have achieved state of the art across countless tasks in natural language processing. These models are highly expressive, comprising at least a hundred million parameters and a dozen layers. Recent evidence suggests that only a few of the final layers need to be fine-tuned for high quality on downstream tasks. Naturally, a subsequent research question is, "how many of the last layers do we need to fine-tune?" In this paper, we precisely answer this question. We examine two recent pretrained language models, BERT and RoBERTa, across standard tasks in textual entailment, semantic similarity, sentiment analysis, and linguistic acceptability. We vary the number of final layers that are fine-tuned, then study the resulting change in task-specific effectiveness. We show that only a fourth of the final layers need to be fine-tuned to achieve 90% of the original quality. Surprisingly, we also find that fine-tuning all layers does not always help.

cs.CL↗

arXiv2019

Explicit Pairwise Word Interaction Modeling Improves Pretrained Transformers for English Semantic Similarity Tasks

In English semantic similarity tasks, classic word embedding-based approaches explicitly model pairwise "interactions" between the word representations of a sentence pair. Transformer-based pretrained language models disregard this notion, instead modeling pairwise word interactions globally and implicitly through their self-attention mechanism. In this paper, we hypothesize that introducing an explicit, constrained pairwise word interaction mechanism to pretrained language models improves their effectiveness on semantic similarity tasks. We validate our hypothesis using BERT on four tasks in semantic textual similarity and answer sentence selection. We demonstrate consistent improvements in quality by adding an explicit pairwise word interaction module to BERT.

cs.CL↗

arXiv2019

Multi-Stage Document Ranking with BERT

The advent of deep neural networks pre-trained via language modeling tasks has spurred a number of successful applications in natural language processing. This work explores one such popular model, BERT, in the context of document ranking. We propose two variants, called monoBERT and duoBERT, that formulate the ranking problem as pointwise and pairwise classification, respectively. These two models are arranged in a multi-stage ranking architecture to form an end-to-end search system. One major advantage of this design is the ability to trade off quality against latency by controlling the admission of candidates into each pipeline stage, and by doing so, we are able to find operating points that offer a good balance between these two competing metrics. On two large-scale datasets, MS MARCO and TREC CAR, experiments show that our model produces results that are either at or comparable to the state of the art. Ablation studies show the contributions of each component and characterize the latency/quality tradeoff space.

cs.IR↗

arXiv2019

Lucene for Approximate Nearest-Neighbors Search on Arbitrary Dense Vectors

We demonstrate three approaches for adapting the open-source Lucene search library to perform approximate nearest-neighbor search on arbitrary dense vectors, using similarity search on word embeddings as a case study. At its core, Lucene is built around inverted indexes of a document collection's (sparse) term-document matrix, which is incompatible with the lower-dimensional dense vectors that are common in deep learning applications. We evaluate three techniques to overcome these challenges that can all be natively integrated into Lucene: the creation of documents populated with fake words, LSH applied to lexical realizations of dense vectors, and k-d trees coupled with dimensionality reduction. Experiments show that the "fake words" approach represents the best balance between effectiveness and efficiency. These techniques are integrated into the Anserini open-source toolkit and made available to the community.

cs.IR↗

arXiv2019

The Performance Envelope of Inverted Indexing on Modern Hardware

This paper explores the performance envelope of "traditional" inverted indexing on modern hardware using the implementation in the open-source Lucene search library. We benchmark indexing throughput on a single high-end multi-core commodity server in a number of configurations varying the media of the source collection and target index, examining a network-attached store, a direct-attached disk array, and an SSD. Experiments show that the largest determinants of performance are the physical characteristics of the source and target media, and that physically isolating the two yields the highest indexing throughput. Results suggest that current indexing techniques have reached physical device limits, and that further algorithmic improvements in performance are unlikely without rethinking the inverted indexing pipeline in light of observed bottlenecks.

cs.IR↗

arXiv2019

Aligning Cross-Lingual Entities with Multi-Aspect Information

Multilingual knowledge graphs (KGs), such as YAGO and DBpedia, represent entities in different languages. The task of cross-lingual entity alignment is to match entities in a source language with their counterparts in target languages. In this work, we investigate embedding-based approaches to encode entities from multilingual KGs into the same vector space, where equivalent entities are close to each other. Specifically, we apply graph convolutional networks (GCNs) to combine multi-aspect information of entities, including topological connections, relations, and attributes of entities, to learn entity embeddings. To exploit the literal descriptions of entities expressed in different languages, we propose two uses of a pretrained multilingual BERT model to bridge cross-lingual gaps. We further propose two strategies to integrate GCN-based and BERT-based modules to boost performance. Extensive experiments on two benchmark datasets demonstrate that our method significantly outperforms existing systems.

cs.CL↗

arXiv2019

Document Expansion by Query Prediction

One technique to improve the retrieval effectiveness of a search engine is to expand documents with terms that are related or representative of the documents' content.From the perspective of a question answering system, this might comprise questions the document can potentially answer. Following this observation, we propose a simple method that predicts which queries will be issued for a given document and then expands it with those predictions with a vanilla sequence-to-sequence model, trained using datasets consisting of pairs of query and relevant documents. By combining our method with a highly-effective re-ranking component, we achieve the state of the art in two retrieval tasks. In a latency-critical regime, retrieval results alone (without re-ranking) approach the effectiveness of more computationally expensive neural re-rankers but are much faster.

cs.IR↗

arXiv2019

Critically Examining the "Neural Hype": Weak Baselines and the Additivity of Effectiveness Gains from Neural Ranking Models

Is neural IR mostly hype? In a recent SIGIR Forum article, Lin expressed skepticism that neural ranking models were actually improving ad hoc retrieval effectiveness in limited data scenarios. He provided anecdotal evidence that authors of neural IR papers demonstrate "wins" by comparing against weak baselines. This paper provides a rigorous evaluation of those claims in two ways: First, we conducted a meta-analysis of papers that have reported experimental results on the TREC Robust04 test collection. We do not find evidence of an upward trend in effectiveness over time. In fact, the best reported results are from a decade ago and no recent neural approach comes close. Second, we applied five recent neural models to rerank the strong baselines that Lin used to make his arguments. A significant improvement was observed for one of the models, demonstrating additivity in gains. While there appears to be merit to neural IR approaches, at least some of the gains reported in the literature appear illusory.

cs.IR↗

arXiv2019

End-to-End Open-Domain Question Answering with BERTserini

We demonstrate an end-to-end question answering system that integrates BERT with the open-source Anserini information retrieval toolkit. In contrast to most question answering and reading comprehension models today, which operate over small amounts of input text, our system integrates best practices from IR with a BERT-based reader to identify answers from a large corpus of Wikipedia articles in an end-to-end fashion. We report large improvements over previous results on a standard benchmark test collection, showing that fine-tuning pretrained BERT with SQuAD is sufficient to achieve high accuracy in identifying answer spans.

cs.CL↗

arXiv2019

The Ubiquity of Large Graphs and Surprising Challenges of Graph Processing: Extended Survey

Graph processing is becoming increasingly prevalent across many application domains. In spite of this prevalence, there is little research about how graphs are actually used in practice. We performed an extensive study that consisted of an online survey of 89 users, a review of the mailing lists, source repositories, and whitepapers of a large suite of graph software products, and in-person interviews with 6 users and 2 developers of these products. Our online survey aimed at understanding: (i) the types of graphs users have; (ii) the graph computations users run; (iii) the types of graph software users use; and (iv) the major challenges users face when processing their graphs. We describe the participants' responses to our questions highlighting common patterns and challenges. Based on our interviews and survey of the rest of our sources, we were able to answer some new questions that were raised by participants' responses to our online survey and understand the specific applications that use graph data and software. Our study revealed surprising facts about graph processing in practice. In particular, real-world graphs represent a very diverse range of entities and are often very large, scalability and visualization are undeniably the most pressing challenges faced by participants, and data integration, recommendations, and fraud detection are very popular applications supported by existing graph software. We hope these findings can guide future research.

cs.DB↗

arXiv2019

Repeatability Corner Cases in Document Ranking: The Impact of Score Ties

Document ranking experiments should be repeatable. However, the interaction between multi-threaded indexing and score ties during retrieval may yield non-deterministic rankings, making repeatability not as trivial as one might imagine. In the context of the open-source Lucene search engine, score ties are broken by internal document ids, which are assigned at index time. Due to multi-threaded indexing, which makes experimentation with large modern document collections practical, internal document ids are not assigned consistently between different index instances of the same collection, and thus score ties are broken unpredictably. This short paper examines the effectiveness impact of such score ties, quantifying the variability that can be attributed to this phenomenon. The obvious solution to this non-determinism and to ensure repeatable document ranking is to break score ties using external collection document ids. This approach, however, comes with measurable efficiency costs due to the necessity of consulting external identifiers during query evaluation.

cs.IR↗

arXiv2019

Simple Attention-Based Representation Learning for Ranking Short Social Media Posts

This paper explores the problem of ranking short social media posts with respect to user queries using neural networks. Instead of starting with a complex architecture, we proceed from the bottom up and examine the effectiveness of a simple, word-level Siamese architecture augmented with attention-based mechanisms for capturing semantic "soft" matches between query and post tokens. Extensive experiments on datasets from the TREC Microblog Tracks show that our simple models not only achieve better effectiveness than existing approaches that are far more complex or exploit a more diverse set of relevance signals, but are also much faster. Implementations of our samCNN (Simple Attention-based Matching CNN) models are shared with the community to support future work.

cs.CL↗

arXiv2019

DocBERT: BERT for Document Classification

We present, to our knowledge, the first application of BERT to document classification. A few characteristics of the task might lead one to think that BERT is not the most appropriate model: syntactic structures matter less for content categories, documents can often be longer than typical BERT input, and documents often have multiple labels. Nevertheless, we show that a straightforward classification model using BERT is able to achieve the state of the art across four popular datasets. To address the computational expense associated with BERT inference, we distill knowledge from BERT-large to small bidirectional LSTMs, reaching BERT-base parity on multiple datasets using 30x fewer parameters. The primary contribution of our paper is improved baselines that can provide the foundation for future work.

cs.CL↗

arXiv2019

Multi-Perspective Relevance Matching with Hierarchical ConvNets for Social Media Search

Despite substantial interest in applications of neural networks to information retrieval, neural ranking models have only been applied to standard ad hoc retrieval tasks over web pages and newswire documents. This paper proposes MP-HCNN (Multi-Perspective Hierarchical Convolutional Neural Network) a novel neural ranking model specifically designed for ranking short social media posts. We identify document length, informal language, and heterogeneous relevance signals as features that distinguish documents in our domain, and present a model specifically designed with these characteristics in mind. Our model uses hierarchical convolutional layers to learn latent semantic soft-match relevance signals at the character, word, and phrase levels. A pooling-based similarity measurement layer integrates evidence from multiple types of matches between the query, the social media post, as well as URLs contained in the post. Extensive experiments using Twitter data from the TREC Microblog Tracks 2011--2014 show that our model significantly outperforms prior feature-based as well and existing neural ranking models. To our best knowledge, this paper presents the first substantial work tackling search over social media posts using neural ranking models.

cs.IR↗

arXiv2019

The Simplest Thing That Can Possibly Work: Pseudo-Relevance Feedback Using Text Classification

Motivated by recent commentary that has questioned today's pursuit of ever-more complex models and mathematical formalisms in applied machine learning and whether meaningful empirical progress is actually being made, this paper tries to tackle the decades-old problem of pseudo-relevance feedback with "the simplest thing that can possibly work". I present a technique based on training a document relevance classifier for each information need using pseudo-labels from an initial ranked list and then applying the classifier to rerank the retrieved documents. Experiments demonstrate significant improvements across a number of newswire collections, with initial rankings supplied by "bag of words" BM25 as well as from a well-tuned query expansion model. While this simple technique draws elements from several well-known threads in the literature, to my knowledge this exact combination has not previously been proposed and evaluated.

cs.IR↗

arXiv2019

Data Augmentation for BERT Fine-Tuning in Open-Domain Question Answering

Recently, a simple combination of passage retrieval using off-the-shelf IR techniques and a BERT reader was found to be very effective for question answering directly on Wikipedia, yielding a large improvement over the previous state of the art on a standard benchmark dataset. In this paper, we present a data augmentation technique using distant supervision that exploits positive as well as negative examples. We apply a stage-wise approach to fine tuning BERT on multiple datasets, starting with data that is "furthest" from the test data and ending with the "closest". Experimental results show large gains in effectiveness over previous approaches on English QA datasets, and we establish new baselines on two recent Chinese QA datasets.

cs.CL↗