AIConfPaper
← Search

Dennis Wei

41 accepted papers

2026

CoFrGeNet: Continued Fraction Architectures for Language Generation

Amit Dhurandhar, Vijil Chenthamarakshan, Dennis Wei, Tejaswini Pedapati, Karthikeyan Natesan Ramamurthy, Rahul Nair
ICML 2026poster

Transformers are arguably the preferred architecture for language generation. In this paper, inspired by continued fractions, we introduce a new function class for generative modeling. The architecture family implementing this function class is named CoFrGeNets - Continued Fraction Generative Networ…

Cited by 0Source ↗Scholar ↗
2026

Dropping Just a Handful of Preferences Can Change Top Large Language Model Rankings

Jenny Y. Huang, Yunyi Shen, Dennis Wei, Tamara Broderick
ICLR 2026poster

We propose a method for evaluating the robustness of widely used LLM ranking systems---variants of a Bradley--Terry model---to dropping a worst-case very small fraction of preference data. Our approach is computationally fast and easy to adopt. When we apply our method to matchups from popular LLM r…

Cited by 0Source ↗code ↗Scholar ↗
2026

Entropy-Aware On-Policy Distillation of Language Models

Woogyeol Jin, Taywon Min, Yongjin Yang, Swanand Kadhe, Yi Zhou, Dennis Wei +2
ICML 2026poster

On-policy distillation is a promising approach for transferring knowledge between language models, where a student learns from dense token-level signals along its own trajectories. This framework typically uses reverse KL divergence, encouraging the student to match the teacher's high-confidence pre…

Cited by 0Source ↗Scholar ↗
2026

Parameterized Abstract Interpretation for Transformer Verification

Pei Huang, Dennis Wei, Omri Isac, Haoze Wu, Min Wu, Clark Barrett
AAAI 2026technical

Transformers based on the self-attention mechanism have become foundational models across a wide range of domains, thereby creating an urgent need for effective formal verification techniques to better understand their behavior and ensure safety guarantees. In this paper, we propose two parameterize

Cited by 0Source ↗PDF ↗Scholar ↗
2025

Fair Continuous Resource Allocation with Equality of Impact

Blossom Metevier, Dennis Wei, Karthikeyan Natesan Ramamurthy, Philip S. Thomas
NeurIPS 2025poster

Recent works have studied fair resource allocation in social settings, where fairness is judged by the impact of allocation decisions rather than more traditional minimum or maximum thresholds on the allocations themselves. Our work significantly adds to this literature by developing continuous reso…

Cited by 0Source ↗Scholar ↗
2025

Final-Model-Only Data Attribution with a Unifying View of Gradient-Based Methods

Dennis Wei, Inkit Padhi, Soumya Ghosh, Amit Dhurandhar, Karthikeyan Natesan Ramamurthy, Maria Chang
NeurIPS 2025poster

Training data attribution (TDA) is concerned with understanding model behavior in terms of the training data. This paper draws attention to the common setting where one has access only to the final trained model, and not the training algorithm or intermediate information from training. We reframe th…

Cited by 0Source ↗Scholar ↗
2025

Invariance Makes LLM Unlearning Resilient Even to Unanticipated Downstream Fine-Tuning

Changsheng Wang, Yihua Zhang, Jinghan Jia, Parikshit Ram, Dennis Wei, Yuguang Yao +3
ICML 2025poster

Machine unlearning presents a promising approach to mitigating privacy and safety concerns in large language models (LLMs) by enabling the selective removal of targeted data or knowledge while preserving model utility. However, existing unlearning methods remain over-sensitive to downstream fine-tun…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2025

Multi-Level Explanations for Generative Language Models

Lucas Monteiro Paes, Dennis Wei, Hyo Jin Do, Hendrik Strobelt, Ronny Luss, Amit Dhurandhar +5
ACL 2025long

Despite the increasing use of large language models (LLMs) for context-grounded tasks like summarization and question-answering, understanding what makes an LLM produce a certain response is challenging. We propose Multi-Level Explanations for Generative Language Models (MExGen), a technique to prov…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2025

Reasoning Model Unlearning: Forgetting Traces, Not Just Answers, While Preserving Reasoning Skills

Changsheng Wang, Chongyu Fan, Yihua Zhang, Jinghan Jia, Dennis Wei, Parikshit Ram +2
EMNLP 2025

Recent advances in large reasoning models (LRMs) have enabled strong multi-step reasoning capabilities. However, existing machine unlearning algorithms are tailored to standard language modeling and fail to address the unique challenges posed by LRMs. In this work, we present the first systematic st

Cited by 0Source ↗PDF ↗Scholar ↗
2024

Causal Bandits with General Causal Models and Interventions

Zirui Yan, Dennis Wei, Dmitriy A Katz, Prasanna Sattigeri, Ali Tajer
AISTATS 2024poster

This paper considers causal bandits (CBs) for the sequential design of interventions in a causal system. The objective is to optimize a reward function via minimizing a measure of cumulative regret with respect to the best sequence of interventions in hindsight. The paper advances the results on CBs…

Cited by 4Source ↗PDF ↗Scholar ↗
2024

Interventional Causal Discovery in a Mixture of DAGs

Burak Varıcı, Dmitriy A Katz, Dennis Wei, Prasanna Sattigeri, Ali Tajer
NeurIPS 2024poster

Causal interactions among a group of variables are often modeled by a single causal graph. In some domains, however, these interactions are best described by multiple co-existing causal graphs, e.g., in dynamical systems or genomics. This paper addresses the hitherto unknown role of interventions in…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2024

SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation

Chongyu Fan, Jiancheng Liu, Yihua Zhang, Eric Wong, Dennis Wei, Sijia Liu
ICLR 2024spotlight

With evolving data regulations, machine unlearning (MU) has become an important tool for fostering trust and safety in today's AI models. However, existing MU methods focusing on data and/or weight perspectives often suffer limitations in unlearning accuracy, stability, and cross-domain applicabilit…

Cited by 130Source ↗PDF ↗code ↗Scholar ↗
2024

Trust Regions for Explanations via Black-Box Probabilistic Certification

Amit Dhurandhar, Swagatam Haldar, Dennis Wei, Karthikeyan Natesan Ramamurthy
ICML 2024poster

Given the black box nature of machine learning models, a plethora of explainability methods have been developed to decipher the factors behind individual decisions. In this paper, we introduce a novel problem of black box (probabilistic) explanation certification. We ask the question: Given a black…

Cited by 2Source ↗PDF ↗code ↗Scholar ↗
2024

Using Causal Inference to Investigate Contraceptive Discontinuation in Sub-Saharan Africa

Victor Akinwande, Megan MacGregor, Celia Cintas, Ehud Karavani, Dennis Wei, Kush R. Varshney +1
IJCAI 2024poster

Discontinuation rates vary by family planning method and across socio-economic contexts. Understanding these variations and their causes is paramount for developing and implementing policies aimed at curbing discontinuation rates. Randomized controlled trials (RCTs) are ideal for obtaining this info…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2023

Convex Bounds on the Softmax Function with Applications to Robustness Verification

Dennis Wei, Haoze Wu, Min Wu, Pin-Yu Chen, Clark Barrett, Eitan Farchi
AISTATS 2023poster

The softmax function is a ubiquitous component at the output of neural networks and increasingly in intermediate layers as well. This paper provides convex lower bounds and concave upper bounds on the softmax function, which are compatible with convex optimization formulations for characterizing neu…

Cited by 13Source ↗PDF ↗code ↗Scholar ↗
2023

Effective Human-AI Teams via Learned Natural Language Rules and Onboarding

Hussein Mozannar, Jimin J Lee, Dennis Wei, Prasanna Sattigeri, Subhro Das, David Sontag
NeurIPS 2023spotlight

People are relying on AI agents to assist them with various tasks. The human must know when to rely on the agent, collaborate with the agent, or ignore its suggestions. In this work, we propose to learn rules grounded in data regions and described in natural language that illustrate how the human sh…

Cited by 20Source ↗PDF ↗code ↗Scholar ↗
2023

Heavy Sets with Applications to Interpretable Machine Learning Diagnostics

Dmitry Malioutov, Sanjeeb Dash, Dennis Wei
AISTATS 2023poster

ML models take on a new life after deployment and raise a host of new challenges: data drift, model recalibration and monitoring. If performance erodes over time, engineers in charge may ask what changed – did the data distribution change, did the model get worse after retraining? We propose a flexi…

Cited by 2Source ↗PDF ↗code ↗Scholar ↗
2023

Interpretable differencing of machine learning models

Swagatam Haldar, Diptikalyan Saha, Dennis Wei, Rahul Nair, Elizabeth M. Daly
UAI 2023poster

Understanding the differences between machine learning (ML) models is of interest in scenarios ranging from choosing amongst a set of competing models, to updating a deployed model with new training data. In these cases, we wish to go beyond differences in overall metrics such as accuracy to identif…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2023

Who Should Predict? Exact Algorithms For Learning to Defer to Humans

Hussein Mozannar, Hunter Lang, Dennis Wei, Prasanna Sattigeri, Subhro Das, David Sontag
AISTATS 2023poster

Automated AI classifiers should be able to defer the prediction to a human decision maker to ensure more accurate predictions. In this work, we jointly train a classifier with a rejector, which decides on each data point whether the classifier or the human should predict. We show that prior approach…

Cited by 57Source ↗PDF ↗code ↗Scholar ↗
2022

On the Safety of Interpretable Machine Learning: A Maximum Deviation Approach

Dennis Wei, Rahul Nair, Amit Dhurandhar, Kush R. Varshney, Elizabeth M. Daly, Moninder Singh
NeurIPS 2022accept

Interpretable and explainable machine learning has seen a recent surge of interest. We focus on safety as a key motivation behind the surge and make the relationship between interpretability and safety more quantitative. Toward assessing safety, we introduce the concept of *maximum deviation* via an…

Cited by 9Source ↗PDF ↗Scholar ↗
2022

Your fairness may vary: Pretrained language model fairness in toxic text classification

Ioana Baldini, Dennis Wei, Karthikeyan Natesan Ramamurthy, Moninder Singh, Mikhail Yurochkin
ACL 2022findings

The popularity of pretrained language models in natural language processing systems calls for a careful evaluation of such models in down-stream tasks, which have a higher potential for societal impact. The evaluation of such systems usually focuses on accuracy measures. Our findings in this paper c…

Cited by 72Source ↗PDF ↗Scholar ↗
2021

CoFrNets: Interpretable Neural Architecture Inspired by Continued Fractions

Isha Puri, Amit Dhurandhar, Tejaswini Pedapati, Karthikeyan Shanmugam, Dennis Wei, Kush R. Varshney
NeurIPS 2021poster

In recent years there has been a considerable amount of research on local post hoc explanations for neural networks. However, work on building interpretable neural architectures has been relatively sparse. In this paper, we present a novel neural architecture, CoFrNet, inspired by the form of contin…

Cited by 13Source ↗PDF ↗Scholar ↗
2021

Conditionally independent data generation

Kartik Ahuja, Prasanna Sattigeri, Karthikeyan Shanmugam, Dennis Wei, Karthikeyan Natesan Ramamurthy, Murat Kocaoglu
UAI 2021poster

Conditional independence (CI) is a fundamental concept with wide applications in machine learning and causal inference. Although the problems of testing CI and estimating divergences have been extensively studied, the complementary problem of generating data that satisfies CI has received much less…

Cited by 5Source ↗PDF ↗Scholar ↗
2021

Treatment Effect Estimation Using Invariant Risk Minimization

Abhin Shah, Kartik Ahuja, Karthikeyan Shanmugam, Dennis Wei, Kush R. Varshney, Amit Dhurandhar
ICASSP 2021accepted

Inferring causal individual treatment effect (ITE) from observational data is a challenging problem whose difficulty is exacerbated by the presence of treatment assignment bias. In this work, we propose a new way to estimate the ITE using the domain generalization framework of invariant risk minimiz…

Cited by 0Source ↗Scholar ↗
2021

What Changed? Interpretable Model Comparison

Rahul Nair, Massimiliano Mattetti, Elizabeth Daly, Dennis Wei, Oznur Alkan, Yunfeng Zhang
IJCAI 2021poster

We consider the problem of distinguishing two machine learning (ML) models built for the same task in a human-interpretable way. As models can fail or succeed in different ways, classical accuracy metrics may mask crucial qualitative differences. This problem arises in a few contexts. In business ap…

Cited by 12Source ↗PDF ↗code ↗Scholar ↗
2020

Characterization of Overlap in Observational Studies

Michael Oberst, Fredrik Johansson, Dennis Wei, Tian Gao, Gabriel Brat, David Sontag +1
AISTATS 2020poster

Overlap between treatment groups is required for non-parametric estimation of causal effects. If a subgroup of subjects always receives the same intervention, we cannot estimate the effect of intervention changes on that subgroup without further assumptions. When overlap does not hold globally, ch…

Cited by 40Source ↗PDF ↗code ↗Scholar ↗
2020

DAGs with No Fears: A Closer Look at Continuous Optimization for Learning Bayesian Networks

Dennis Wei, Tian Gao, Yue Yu
NeurIPS 2020spotlight

This paper re-examines a continuous optimization framework dubbed NOTEARS for learning Bayesian networks. We first generalize existing algebraic characterizations of acyclicity to a class of matrix polynomials. Next, focusing on a one-parameter-per-edge setting, it is shown that the Karush-Kuhn-Tuck…

Cited by 91Source ↗PDF ↗code ↗Scholar ↗
2020

Is There a Trade-Off Between Fairness and Accuracy? A Perspective Using Mismatched Hypothesis Testing

Sanghamitra Dutta, Dennis Wei, Hazar Yueksel, Pin-Yu Chen, Sijia Liu, Kush Varshney
ICML 2020poster

A trade-off between accuracy and fairness is almost taken as a given in the existing literature on fairness in machine learning. Yet, it is not preordained that accuracy should decrease with increased fairness. Novel to this work, we examine fair classification through the lens of mismatched hypothe…

Cited by 195Source ↗PDF ↗Scholar ↗
2020

Optimized Score Transformation for Fair Classification

Dennis Wei, Karthikeyan Natesan Ramamurthy, Flavio Calmon
AISTATS 2020poster

This paper considers fair probabilistic classification where the outputs of primary interest are predicted probabilities, commonly referred to as scores. We formulate the problem of transforming scores to satisfy fairness constraints while minimizing the loss in utility. The formulation can be appli…

Cited by 56Source ↗PDF ↗Scholar ↗
2018

On the Supermodularity of Active Graph-Based Semi-Supervised Learning with Stieltjes Matrix Regularization

Pin-Yu Chen, Dennis Wei
ICASSP 2018accepted

Active graph-based semi-supervised learning (AG-SSL) aims to select a small set of labeled examples and utilize their graph-based relation to other unlabeled examples to aid in machine learning tasks. It is also closely related to the sampling theory in graph signal processing. In this paper, we rev…

Cited by 0Source ↗Scholar ↗
2017

Optimized Pre-Processing for Discrimination Prevention

Flavio Calmon, Dennis Wei, Bhanukiran Vinzamuri, Karthikeyan Natesan Ramamurthy, Kush R Varshney
NeurIPS 2017poster

Non-discrimination is a recognized objective in algorithmic decision making. In this paper, we introduce a novel probabilistic formulation of data pre-processing for reducing discrimination. We propose a convex optimization for learning a data transformation with three goals: controlling discriminat…

Cited by 1131Source ↗PDF ↗Scholar ↗
2016

Empirically-estimable multi-class classification bounds

Alan Wisler, Visar Berisha, Dennis Wei, Karthikeyan Ramamurthy, Andreas Spanias
ICASSP 2016accepted

In this paper, we extend previously developed non-parametric bounds on the Bayes risk in binary classification problems to multi-class problems. In comparison with the well-known Bhattacharyya bound which is typically calculated by employing parametric assumptions, the bounds proposed in this paper…

Cited by 0Source ↗Scholar ↗