72 citations · 328 across the 29 of their papers we have counts for
30 papers
AMNESIA: A Large Scale Medical Unlearning Benchmark Suite with Disease-Informed Analysis
Saeedeh Davoudi, Reihaneh Iranmanesh, Ophir Frieder +1
Medical knowledge is continuously evolving. This creates a need to update or selectively forget information encoded in already-trained medical LLMs. Machine unlearning aims to remo…
TARAZ: Persian Short-Answer Question Benchmark for Cultural Evaluation of Language Models
Reihaneh Iranmanesh, Saeedeh Davoudi, Pasha Abrishamchian +2
This paper presents a comprehensive evaluation framework for assessing the cultural competence of large language models (LLMs) in Persian. Existing Persian cultural benchmarks rely…
A Picture of Agentic Search
Francesca Pezzuti, Ophir Frieder, Fabrizio Silvestri +2
With automated systems increasingly issuing search queries alongside humans, Information Retrieval (IR) faces a major shift. Yet IR remains human-centred, with systems, evaluation…
DRAMA: Domain Retrieval using Adaptive Module Allocation
Pranav Kasela, Marco Braga, Ophir Frieder +3
Neural models are increasingly used in Web-scale Information Retrieval (IR). However, relying on these models introduces substantial computational and energy requirements, leading…
Dagstuhl Perspectives Workshop 24352 -- Conversational Agents: A Framework for Evaluation (CAFE): Manifesto
Christine Bauer, Li Chen, Nicola Ferro +19
During the workshop, we deeply discussed what CONversational Information ACcess (CONIAC) is and its unique features, proposing a world model abstracting it, and defined the Convers…
Scaling Electronic Health Record Foundation Models for Population Health Management
Liwen Sun, Hao-Ren Yao, Ophir Frieder +2
Population health management requires scalable methods to identify individuals at risk of chronic diseases such as cardiovascular conditions and cancer, yet existing approaches rel…
GRIT: Graph-based Recall Improvement for Task-oriented E-commerce Queries
Hrishikesh Kulkarni, Surya Kallumadi, Sean MacAvaney +2
Many e-commerce search pipelines have four stages, namely: retrieval, filtering, ranking, and personalized-reranking. The retrieval stage must be efficient and yield high recall be…
LexBoost: Improving Lexical Document Retrieval with Nearest Neighbors
Hrishikesh Kulkarni, Nazli Goharian, Ophir Frieder +1
Sparse retrieval methods like BM25 are based on lexical overlap, focusing on the surface form of the terms that appear in the query and the document. The use of inverted indices in…
Genetic Approach to Mitigate Hallucination in Generative IR
Hrishikesh Kulkarni, Nazli Goharian, Ophir Frieder +1
Generative language models hallucinate. That is, at times, they generate factually flawed responses. These inaccuracies are particularly insidious because the responses are fluent…
Lexically-Accelerated Dense Retrieval
Hrishikesh Kulkarni, Sean MacAvaney, Nazli Goharian +1
Retrieval approaches that score documents based on learned dense vectors (i.e., dense retrieval) rather than lexical signals (i.e., conventional retrieval) are increasingly popular…
Caching Historical Embeddings in Conversational Search
Ophir Frieder, Ida Mele, Cristina Ioana Muntean +3
Rapid response, namely low latency, is fundamental in search applications; it is particularly so in interactive search sessions, such as those encountered in conversational setting…
Self-supervised Representation Learning on Electronic Health Records with Graph Kernel Infomax
Hao-Ren Yao, Nairen Cao, Katina Russell +3
Learning Electronic Health Records (EHRs) representation is a preeminent yet under-discovered research topic. It benefits various clinical decision support applications, e.g., medi…
TAR on Social Media: A Framework for Online Content Moderation
Eugene Yang, David D. Lewis, Ophir Frieder
Content moderation (removing or limiting the distribution of posts based on their contents) is one tool social networks use to fight problems such as harassment and disinformation.…
Certifying One-Phase Technology-Assisted Reviews
David D. Lewis, Eugene Yang, Ophir Frieder
Technology-assisted review (TAR) workflows based on iterative active learning are widely used in document review applications. Most stopping rules for one-phase TAR workflows lack…
Heuristic Stopping Rules For Technology-Assisted Review
Eugene Yang, David D. Lewis, Ophir Frieder
Technology-assisted review (TAR) refers to human-in-the-loop active learning workflows for finding relevant documents in large collections. These workflows often must meet a target…
On Minimizing Cost in Legal Document Review Workflows
Eugene Yang, David D. Lewis, Ophir Frieder
Technology-assisted review (TAR) refers to human-in-the-loop machine learning workflows for document review in legal discovery and other high recall review tasks. Attorneys and leg…
Goldilocks: Just-Right Tuning of BERT for Technology-Assisted Review
Eugene Yang, Sean MacAvaney, David D. Lewis +1
Technology-assisted review (TAR) refers to iterative active learning workflows for document review in high recall retrieval (HRR) tasks. TAR research and most commercial TAR softwa…
The Analysis from Nonlinear Distance Metric to Kernel-based Drug Prescription Prediction System
Der-Chen Chang, Ophir Frieder, Chi-Feng Hung +1
Distance metrics and their nonlinear variant play a crucial role in machine learning based real-world problem solving. We demonstrated how Euclidean and cosine distance measures di…
Cross-Global Attention Graph Kernel Network Prediction of Drug Prescription
Hao-Ren Yao, Der-Chen Chang, Ophir Frieder +3
We present an end-to-end, interpretable, deep-learning architecture to learn a graph kernel that predicts the outcome of chronic disease drug prescription. This is achieved through…
GUIR at SemEval-2020 Task 12: Domain-Tuned Contextualized Models for Offensive Language Detection
Sajad Sotudeh, Tong Xiang, Hao-Ren Yao +4
Offensive language detection is an important and challenging task in natural language processing. We present our submissions to the OffensEval 2020 shared task, which includes thre…
Efficient Document Re-Ranking for Transformers by Precomputing Term Representations
Sean MacAvaney, Franco Maria Nardini, Raffaele Perego +3
Deep pretrained transformer networks are effective at various ranking tasks, such as question answering and ad-hoc document ranking. However, their computational expenses deem them…
Training Curricula for Open Domain Answer Re-Ranking
Sean MacAvaney, Franco Maria Nardini, Raffaele Perego +3
In precision-oriented tasks like answer ranking, it is more important to rank many relevant answers highly than to retrieve all relevant answers. It follows that a good ranking str…
Expansion via Prediction of Importance with Contextualization
Sean MacAvaney, Franco Maria Nardini, Raffaele Perego +3
The identification of relevance with little textual context is a primary challenge in passage retrieval. We address this problem with a representation-based ranking approach that:…
Interaction Matching for Long-Tail Multi-Label Classification
Sean MacAvaney, Franck Dernoncourt, Walter Chang +2
We present an elegant and effective approach for addressing limitations in existing multi-label classification models by incorporating interaction matching, a concept shown to be u…