Home on Abhiraj Bibhar

FedCare: Privacy-Preserving Clinical Language Models

Mon, 01 Jan 0001 00:00:00 +0000

A multi-institution collaboration building federated NLP models for clinical decision support without centralizing patient data.

Trustworthy & Private Machine Learning

Mon, 01 Jan 0001 00:00:00 +0000

Developing federated learning and differential privacy techniques that let sensitive data — especially in healthcare — be used for training without compromising individual privacy.

GlassBox: Visual Debugging for Graph Neural Networks

Mon, 01 Jan 0001 00:00:00 +0000

An open-source toolkit for interactively inspecting attention and message-passing behavior in GNNs.

Human-Centered AI

Mon, 01 Jan 0001 00:00:00 +0000

Studying how people build trust and mental models of AI systems, and designing interaction techniques that support meaningful human oversight.

EquiTutor: Adaptive Learning at Scale

Mon, 01 Jan 0001 00:00:00 +0000

A large-scale study of multimodal tutoring interfaces deployed across three partner school districts.

Explainable & Interpretable Systems

Mon, 01 Jan 0001 00:00:00 +0000

Creating visualization and analysis tools that make complex models — including graph neural networks and large language models — easier to understand and debug.

AI for Education

Mon, 01 Jan 0001 00:00:00 +0000

Building adaptive tutoring systems that respond to learners’ cognitive load and support equitable access to personalized instruction.

Our federated NLP paper has been accepted to JMLR

Fri, 12 Jun 2026 00:00:00 +0000

Our work on attention-guided federated learning for privacy-preserving clinical NLP has been accepted for publication in the Journal of Machine Learning Research.

Attention-Guided Federated Learning for Privacy-Preserving Clinical NLP

Mon, 01 Jun 2026 00:00:00 +0000

Abstract

Training clinical natural language processing models typically requires centralizing patient records, which is often legally or ethically infeasible across institutions. We introduce an attention-guided federated learning protocol that allows clinical language models to be trained collaboratively across multiple hospital systems without any patient data leaving its host institution.

Why Federated Learning Is Harder Than the Tutorials Make It Look

Mon, 18 May 2026 00:00:00 +0000

The gap between a federated averaging demo notebook and a system that survives real hospitals, spotty networks, and adversarial clients is bigger than most tutorials let on.

Lab awarded continuation funding from the NSF

Fri, 03 Apr 2026 00:00:00 +0000

The National Science Foundation has renewed support for our CAREER project on trustworthy federated learning through 2026.

What I Changed After a Decade of Teaching Algorithms

Mon, 02 Mar 2026 00:00:00 +0000

Ten years of CS 210 taught me that the biggest predictor of student success isn’t talent — it’s how quickly wrong mental models get corrected.

CHI paper receives Honorable Mention

Thu, 22 Jan 2026 00:00:00 +0000

Our paper on human-in-the-loop interfaces for trustworthy AI decision support received an Honorable Mention award at ACM CHI.

Explainability Is Not a Feature You Bolt On Afterward

Wed, 14 Jan 2026 00:00:00 +0000

Post-hoc explanation methods are useful, but treating explainability as a wrapper around an already-trained black box quietly limits what kinds of trust you can actually earn.

Human-in-the-Loop Interfaces for Trustworthy AI Decision Support

Thu, 01 May 2025 00:00:00 +0000

Abstract

We report on a mixed-methods study of 48 clinicians interacting with three AI-assisted diagnostic interfaces that varied in how they presented model confidence and supporting evidence. Confidence calibration displays significantly improved appropriate reliance on model suggestions, and clinicians consistently preferred interfaces that surfaced underlying evidence rather than raw confidence scores. Time pressure substantially altered how much clinicians engaged with explanatory content — a factor largely absent from prior lab-based evaluations of AI decision support.

A Survey of Explainability Methods for Graph Neural Networks

Sat, 01 Feb 2025 00:00:00 +0000

Abstract

Graph neural networks (GNNs) are increasingly deployed in high-stakes settings, yet the tools available for explaining their predictions lag behind those for standard deep networks. This survey organizes the growing landscape of GNN explainability methods along three axes — instance-level versus model-level, perturbation-based versus gradient-based, and faithfulness versus plausibility — and evaluates each family against a consistent set of faithfulness benchmarks.

Low-Resource Adaptation of Large Language Models for Regional Dialects

Fri, 01 Nov 2024 00:00:00 +0000

Abstract

Large language models perform unevenly across regional dialects that are underrepresented in web-scale training corpora. We propose a parameter-efficient adaptation method that fine-tunes a small set of dialect-specific adapter layers using as few as 5,000 labeled examples, substantially closing the performance gap on downstream tasks including sentiment analysis, named entity recognition, and machine translation for four regional dialect groups.

Fairness-Aware Ranking in Multi-Stakeholder Recommender Systems

Sat, 01 Jun 2024 00:00:00 +0000

Abstract

Recommender systems increasingly serve multiple stakeholders — consumers, item providers, and platform operators — whose interests are not always aligned. We formalize a fairness-aware ranking objective that balances consumer relevance against provider-side exposure fairness, and propose an efficient re-ranking algorithm that achieves Pareto-improving trade-offs relative to relevance-only baselines on three large-scale recommendation datasets.

Toward Reproducible Benchmarks in Federated Learning Research

Fri, 01 Dec 2023 00:00:00 +0000

Abstract

Federated learning papers frequently report results on incompatible experimental setups — differing client counts, non-IID partitioning strategies, and communication budgets — making cross-paper comparison unreliable. This position paper proposes a minimal standardized benchmark protocol and reports a re-evaluation of six widely cited federated learning methods under matched conditions, finding that reported performance gaps shrink substantially once experimental setup is controlled for.

Interactive Visualization Techniques for Deep Model Debugging

Sun, 01 Oct 2023 00:00:00 +0000

Abstract

Debugging deep learning models remains largely a manual, intuition-driven process. We present GlassBox, an interactive visualization system that lets researchers inspect layer activations, attention weights, and gradient flow simultaneously across a model’s forward and backward passes, with linked views that highlight where behavior diverges from expectations.

Cognitive Load in Multimodal Tutoring Systems

Thu, 01 Sep 2022 00:00:00 +0000

Abstract

Multimodal tutoring systems that combine text, audio, and visual explanation can improve learning outcomes, but they can also increase extraneous cognitive load if poorly designed. Across three classroom studies with a combined 240 students, we measure cognitive load using dual-task response times and find that adaptive modality selection — choosing which modality to emphasize based on a learner’s real-time performance — reduces extraneous load without harming learning gains.

About

Mon, 01 Jan 0001 00:00:00 +0000

Abhiraj Bibhar is a Professor of Computer Science at Riverbend University, where she directs the Human-Centered Machine Learning Lab. Her research sits at the intersection of machine learning, human-computer interaction, and privacy, with a focus on building AI systems that are trustworthy, explainable, and genuinely useful to the people who rely on them.

Contact

Mon, 01 Jan 0001 00:00:00 +0000

Curriculum Vitae

Mon, 01 Jan 0001 00:00:00 +0000

Data Structures & Algorithms

Mon, 01 Jan 0001 00:00:00 +0000

Fundamental data structures, algorithm design paradigms, and complexity analysis with hands-on programming labs.

Human-Computer Interaction

Mon, 01 Jan 0001 00:00:00 +0000

Design and evaluation of interactive systems, covering usability, accessibility, and participatory design methods.

Machine Learning Foundations

Mon, 01 Jan 0001 00:00:00 +0000

Core principles of statistical learning, optimization, and neural networks with an emphasis on rigorous evaluation and reproducibility.

Page Not Found

Mon, 01 Jan 0001 00:00:00 +0000

Search

Mon, 01 Jan 0001 00:00:00 +0000

Seminar in Trustworthy AI

Mon, 01 Jan 0001 00:00:00 +0000

A research seminar examining fairness, privacy, and explainability in modern machine learning systems through primary literature.