Search Results for author:
Found 46 papers, 15 papers with code
The Surprising Effectiveness of Rankers Trained on Expanded Queries
In our extensive experiments on the DL-Hard dataset, we find that a principled query performance based scoring method using base and specialized ranker offers a significant improvement of up to 25% on the passage ranking task and up to 48. 4% on the document ranking task when compared to the baseline performance of using original queries, even outperforming SOTA model.
NUMTEMP: A real-world benchmark to verify claims with statistical and temporal expressions
This addresses the challenge of verifying real-world numerical claims, which are complex and often lack precise information, not addressed by existing works that mainly focus on synthetic claims.
RankingSHAP -- Listwise Feature Attribution Explanations for Ranking Models
We evaluate RankingSHAP for commonly used learning-to-rank datasets to showcase the more nuanced use of an attribution method while highlighting the limitations of selection-based explanations.
Temporal Blind Spots in Large Language Models
In this study, we aim to investigate the underlying limitations of general-purpose LLMs when deployed for tasks that require a temporal understanding.
Data Augmentation for Sample Efficient and Robust Document Ranking
We then adapt a family of contrastive losses for the document ranking task that can exploit the augmented data to learn an effective ranking model.
Efficient Neural Ranking using Forward Indexes and Lightweight Encoders
Dual-encoder-based dense retrieval models have become the standard in IR.
In-Context Ability Transfer for Question Decomposition in Complex QA
We transfer the ability to decompose complex questions to simpler questions or generate step-by-step rationales to LLMs, by careful selection from available data sources of related tasks.
DINE: Dimensional Interpretability of Node Embeddings
Towards that, we first develop new metrics that measure the global interpretability of embedding vectors based on the marginal contribution of the embedding dimensions to predicting graph structure.
Context Aware Query Rewriting for Text Rankers using LLM
We find that there are two inherent limitations of using LLMs as query re-writers -- concept drift when using only queries as prompts and large inference costs during query processing.
Query Understanding in the Age of Large Language Models
Querying, conversing, and controlling search and information-seeking interfaces using natural language are fast becoming ubiquitous with the rise and adoption of large-language models (LLM).
The Effect of Masking Strategies on Knowledge Retention by Language Models
To foster reproducibility, the code, as well as the data used in this paper, are openly available.
Distribution-Aligned Fine-Tuning for Efficient Neural Retrieval
Dual-encoder-based neural retrieval models achieve appreciable performance and complement traditional lexical retrievers well due to their semantic matching capabilities, which makes them a common choice for hybrid IR systems.
Explainable Information Retrieval: A Survey
Explainable information retrieval is an emerging research area aiming to make transparent and trustworthy information retrieval systems.
Supervised Contrastive Learning Approach for Contextual Ranking
This paper proposes a simple yet effective method to improve ranking performance on smaller datasets using supervised contrastive learning for the document ranking problem.
BAGEL: A Benchmark for Assessing Graph Neural Network Explanations
Given a GNN model, several interpretability approaches exist to explain GNN models with diverse (sometimes conflicting) evaluation methodologies.
BERT Rankers are Brittle: a Study using Adversarial Document Perturbations
The aim of our algorithms is to add/replace a small number of tokens to a highly relevant or non-relevant document to cause a large rank demotion or promotion.
SparCAssist: A Model Risk Assessment Assistant Based on Sparse Generated Counterfactuals
We introduce SparcAssist, a general-purpose risk assessment tool for the machine learning models trained for language tasks.
Efficient Neural Ranking using Forward Indexes
In this paper, we propose the Fast-Forward index -- a simple vector forward index that facilitates ranking documents using interpolation of lexical and semantic scores -- as a replacement for contextual re-rankers and dense indexes based on nearest neighbor search.
FaxPlainAC: A Fact-Checking Tool Based on EXPLAINable Models with HumAn Correction in the Loop
It is therefore important to conduct user studies to correct models' inference biases and improve the model in a life-long learning manner in the future according to the user feedback.
Learnt Sparsification for Interpretable Graph Neural Networks
However, GNNs remain hard to interpret as the interplay between node features and graph structure is only implicitly learned.
Extractive Explanations for Interpretable Text Ranking
We introduce the Select-and-Rank paradigm for document ranking, where we first output an explanation as a selected subset of sentences in a document.
Towards Axiomatic Explanations for Neural Ranking Models
We investigate whether one can explain the behavior of neural ranking models in terms of their congruence with well understood principles of document ranking by using established theories from axiomatic IR.
Exploiting Sentence-Level Representations for Passage Ranking
Recently, pre-trained contextual models, such as BERT, have shown to perform well in language related tasks.
BERTnesia: Investigating the capture and forgetting of knowledge in BERT
We found that ranking models forget the least and retain more knowledge in their final layer compared to masked language modeling and question-answering.
Zorro: Valid, Sparse, and Stable Explanations in Graph Neural Networks
In this paper, we lay down some of the fundamental principles that an explanation method for graph neural networks should follow and introduce a metric RDT-Fidelity as a measure of the explanation's effectiveness.
Towards Benchmarking the Utility of Explanations for Model Debugging
Post-hoc explanation methods are an important class of approaches that help understand the rationale underlying a trained model's decision.
An In-depth Analysis of Passage-Level Label Transfer for Contextual Document Ranking
However, the documents are longer than the passages and such document ranking models suffer from the token limitation (512) of BERT.
Revisiting the Auction Algorithm for Weighted Bipartite Perfect Matchings
We study the classical weighted perfect matchings problem for bipartite graphs or sometimes referred to as the assignment problem, i. e., given a weighted bipartite graph $G = (U\cup V, E)$ with weights $w : E \rightarrow \mathcal{R}$ we are interested to find the maximum matching in $G$ with the minimum/maximum weight.
Data Structures and Algorithms Discrete Mathematics Combinatorics
Dissonance Between Human and Machine Understanding
Are humans consistently better at selecting features that make image recognition more accurate?
Explain and Predict, and then Predict Again
A desirable property of learning systems is to be both effective and interpretable.
Hard Masking for Explaining Graph Neural Networks
Graph Neural Networks (GNNs) are a flexible and powerful family of models that build nodes' representations on irregular graph-structured data.
BERTnesia: Investigating the capture and forgetting of knowledge in BERT
We found that ranking models forget the least and retain more knowledge in their final layer.
Conversational Search -- A Report from Dagstuhl Seminar 19461
Dagstuhl Seminar 19461 "Conversational Search" was held on 10-15 November 2019.
Valid Explanations for Learning to Rank Models
In extensive quantitative experiments we show that our approach outperforms other model agnostic explanation approaches across pointwise, pairwise and listwise LTR models in validity while not compromising on completeness.
Question Answering over Curated and Open Web Sources
The last few years have seen an explosion of research on the topic of automated question answering (QA), spanning the communities of information retrieval, natural language processing, and artificial intelligence.
Boilerplate Removal using a Neural Sequence Labeling Model
The extraction of main content from web pages is an important task for numerous applications, ranging from usability aspects, like reader views for news articles in web browsers, to information retrieval or natural language processing.
Finding Interpretable Concept Spaces in Node Embeddings using Knowledge Bases
In this paper we propose and study the novel problem of explaining node embeddings by finding embedded human interpretable subspaces in already trained unsupervised node representation embeddings.
A study on the Interpretability of Neural Retrieval Models using DeepSHAP
In image classification, the reference input tends to be a plain black image.
Understanding, Categorizing and Predicting Semantic Image-Text Relations
Two modalities are often used to convey information in a complementary and beneficial manner, e. g., in online news, videos, educational resources, or scientific publications.
A Comparative Study for Unsupervised Network Representation Learning
However, there is no common ground for systematic comparison of embeddings to understand their behavior for different graphs and tasks.
Asynchronous Training of Word Embeddings for Large Text Corpora
In this paper, we propose a scalable approach to train word embeddings by partitioning the input space instead in order to scale to massive text corpora while not sacrificing the performance of the embeddings.
Node Representation Learning for Directed Graphs
We also unearth the limitations of evaluations on directed graphs in previous works and propose a clear strategy for evaluating link prediction and graph reconstruction in directed graphs.
Fine Grained Citation Span for References in Wikipedia
For a Wikipedia article, determining the \emph{citation span} of a citation, i. e. what content is covered by a citation, is important as it helps decide for which content citations are still missing.
Finding News Citations for Wikipedia
An important editing policy in Wikipedia is to provide citations for added statements in Wikipedia pages, where statements can be arbitrary pieces of text, ranging from a sentence to a paragraph.
Automated News Suggestions for Populating Wikipedia Entity Pages
We propose a two-stage supervised approach for suggesting news articles to entity pages for a given state of Wikipedia.
Balancing Novelty and Salience: Adaptive Learning to Rank Entities for Timeline Summarization of High-impact Events
In this work, we present a novel approach for timeline summarization of high-impact events, which uses entities instead of sentences for summarizing the event at each individual point in time.
