Author name cluster

Huzefa Rangwala

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

20 papers

2 author rows

ICLR Conference 2025 Conference Paper

AgentOccam: A Simple Yet Strong Baseline for LLM-Based Web Agents

Ke Yang 0003
Yao Liu 0009
Sapana Chaudhary
Rasool Fakoor
Pratik Chaudhari
George Karypis
Huzefa Rangwala

Autonomy via agents based on large language models (LLMs) that can carry out personalized yet standardized tasks presents a significant opportunity to drive human efficiency. There is an emerging need and interest in automating web tasks (e.g., booking a hotel for a given date within a budget). Being a practical use case itself, the web agent also serves as an important proof-of-concept example for various agent grounding scenarios, with its success promising advancements in many future applications. Meanwhile, much prior research focuses on handcrafting their web agent strategies (e.g., agent's prompting templates, reflective workflow, role-play and multi-agent systems, search or sampling methods, etc.) and the corresponding in-context examples. However, these custom strategies often struggle with generalizability across all potential real-world applications. On the other hand, there has been limited study on the misalignment between a web agent's observation and action representation, and the data on which the agent's underlying LLM has been pre-trained. This discrepancy is especially notable when LLMs are primarily trained for language completion rather than tasks involving embodied navigation actions and symbolic web elements. In our study, we enhance an LLM-based web agent by simply refining its observation and action space, aligning these more closely with the LLM's capabilities. This approach enables our base agent to significantly outperform previous methods on a wide variety of web tasks. Specifically, on WebArena, a benchmark featuring general-purpose web interaction tasks, our agent AgentOccam surpasses the previous state-of-the-art and concurrent work by 9.8 (+29.4%) and 5.9 (+15.8%) absolute points respectively, and boosts the success rate by 26.6 points (+161%) over similar plain web agents with its observation and action space alignment. Furthermore, on WebVoyager benchmark comprising tasks defined on real-world websites, AgentOccam exceeds the former best agent by 2.4 points (+4.6%) on tasks with deterministic answers. We achieve this without using in-context examples, new agent roles, online feedback or search strategies. AgentOccam's simple design highlights LLMs' impressive zero-shot performance on web tasks, and underlines the critical role of carefully tuning observation and action spaces for LLM-based agents.

Details

ICLR Conference 2025 Conference Paper

AutoG: Towards automatic graph construction from tabular data

Zhikai Chen
Han Xie
Jian Zhang 0113
Xiang Song 0003
Jiliang Tang
Huzefa Rangwala
George Karypis

Recent years have witnessed significant advancements in graph machine learning (GML), with its applications spanning numerous domains. However, the focus of GML has predominantly been on developing powerful models, often overlooking a crucial initial step: constructing suitable graphs from common data formats, such as tabular data. This construction process is fundamental to applying graph-based models, yet it remains largely understudied and lacks formalization. Our research aims to address this gap by formalizing the graph construction problem and proposing an effective solution. We identify two critical challenges to achieve this goal: 1. The absence of dedicated datasets to formalize and evaluate the effectiveness of graph construction methods, and 2. Existing automatic construction methods can only be applied to some specific cases, while tedious human engineering is required to generate high-quality graphs. To tackle these challenges, we present a two-fold contribution. First, we introduce a set of datasets to formalize and evaluate graph construction methods. Second, we propose an LLM-based solution, AutoG, automatically generating high-quality graph schemas without human intervention. The experimental results demonstrate that the quality of constructed graphs is critical to downstream task performance, and AutoG can generate high-quality graphs that rival those produced by human experts. Our code can be accessible from https://github.com/amazon-science/Automatic-Table-to-Graph-Generation.

Details

TMLR Journal 2025 Journal Article

Bridging the Training-Inference Gap in LLMs by Leveraging Self-Generated Tokens

Zhepeng Cen
Yao Liu
Siliang Zeng
Pratik Chaudhari
Huzefa Rangwala
George Karypis
Rasool Fakoor

Language models are often trained to maximize the likelihood of the next token given past tokens in the training dataset. However, during inference time, they are utilized differently, generating text sequentially and auto-regressively by using previously \emph{generated} tokens as input to predict the next one. Marginal differences in predictions at each step can cascade over successive steps, resulting in different distributions from what the models were trained for and potentially leading to unpredictable behavior. This paper proposes two simple approaches based on model own generation to address this discrepancy between the training and inference time. Our first approach is Batch-Scheduled Sampling, where, during training, we stochastically choose between the ground-truth token from the dataset and the model's own generated token as input to predict the next token. This is done in an offline manner, modifying the context window by interleaving ground-truth tokens with those generated by the model. Our second approach is Reference-Answer-based Correction, where we explicitly incorporate a self-correction capability into the model during training. This enables the model to effectively self-correct the gaps between the generated sequences and the ground truth data without relying on an external oracle model. By incorporating our proposed strategies during training, we have observed an overall improvement in performance compared to baseline methods, as demonstrated by our extensive experiments using summarization, general question-answering, and math question-answering tasks.

PDF Details

TMLR Journal 2025 Journal Article

LC-PLM: Long-context Protein Language Modeling Using Bidirectional Mamba with Shared Projection Layers

Yingheng Wang
Zichen Wang
Gil Sadeh
Luca Zancato
Alessandro Achille
George Karypis
Huzefa Rangwala

Self-supervised training of language models (LMs) has seen great success for protein sequences in learning meaningful representations and for generative drug design. Most protein LMs are based on the Transformer architecture trained on individual proteins with short context lengths. Such protein LMs cannot extrapolate to longer proteins and protein complexes well. They also fail to account for the underlying biological mechanisms carried out by biomolecular interactions and dynamics i.e., proteins often interact with other proteins, molecules, and pathways in complex biological systems. In this work, we propose LC-PLM based on an alternative protein LM architecture, BiMamba-S, built upon selective structured state-space models, to learn high-quality universal protein representations at the amino acid token level using masked language modeling. We also introduce its graph-contextual variant, LC-PLM, which contextualizes protein-protein interaction (PPI) graphs for a second stage of training. LC-PLM demonstrates favorable neural scaling laws, better length extrapolation capability, and up to 30% and 16% improvements on protein downstream tasks compared to Transformer-based ESM-2 when trained with 100B and 1T tokens, respectively. LC-PLM-G further trained within the context of PPI graphs shows promising results on protein structure and function prediction tasks. Our study demonstrates the benefit of increasing the context size with computationally efficient LM architecture (e.g., structured state space models) in learning universal protein representations and incorporating molecular interaction contexts contained in biological graphs. Model is available at github.com/amazon-science/LC-PLM.

PDF Details

NeurIPS Conference 2025 Conference Paper

Mitra: Mixed Synthetic Priors for Enhancing Tabular Foundation Models

Xiyuan Zhang
Danielle Maddix Robinson
Junming Yin
Nick Erickson
Abdul Fatir Ansari
Boran Han
Shuai Zhang
Leman Akoglu

Since the seminal work of TabPFN, research on tabular foundation models (TFMs) based on in-context learning (ICL) has challenged long-standing paradigms in machine learning. Without seeing any real-world data, models pretrained on purely synthetic datasets generalize remarkably well across diverse datasets, often using only a moderate number of in-context examples. This shifts the focus in tabular machine learning from model architecture design to the design of synthetic datasets, or, more precisely, to the prior distributions that generate them. Yet the guiding principles for prior design remain poorly understood. This work marks the first attempt to address the gap. We systematically investigate and identify key properties of synthetic priors that allow pretrained TFMs to generalize well. Based on these insights, we introduce Mitra, a TFM trained on a curated mixture of synthetic priors selected for their diversity, distinctiveness, and performance on real-world tabular data. Mitra consistently outperforms state-of-the-art TFMs, such as TabPFNv2 and TabICL, across both classification and regression benchmarks, with better sample efficiency.

PDF Details

NeurIPS Conference 2025 Conference Paper

MLZero: A Multi-Agent System for End-to-end Machine Learning Automation

Haoyang Fang
Boran Han
Nick Erickson
Xiyuan Zhang
Su Zhou
Anirudh Dagar
Jiani Zhang
Ali Caner Turkmen

Existing AutoML systems have advanced the automation of machine learning (ML); however, they still require substantial manual configuration and expert input, particularly when handling multimodal data. We introduce MLZero, a novel multi-agent framework powered by Large Language Models (LLMs) that enables end-to-end ML automation across diverse data modalities with minimal human intervention. A cognitive perception module is first employed, transforming raw multimodal inputs into perceptual context that effectively guides the subsequent workflow. To address key limitations of LLMs, such as hallucinated code generation and outdated API knowledge, we enhance the iterative code generation process with semantic and episodic memory. MLZero demonstrates superior performance on MLE-Bench Lite, outperforming all competitors in both success rate and solution quality, securing six gold medals. Furthermore, when evaluated on our Multimodal AutoML Agent Benchmark, which includes 25 more challenging tasks spanning diverse data modalities, MLZero outperforms the competing methods by a large margin with a success rate of 0. 92 (+263. 6\%) and an average rank of 2. 28. Our approach maintains its robust effectiveness even with a compact 8B LLM, outperforming full-size systems from existing solutions.

PDF Details

TMLR Journal 2025 Journal Article

Offline Learning and Forgetting for Reasoning with Large Language Models

Tianwei Ni
Allen Nie
Sapana Chaudhary
Yao Liu
Huzefa Rangwala
Rasool Fakoor

Leveraging inference-time search in large language models has proven effective in further enhancing a trained model's capability to solve complex mathematical and reasoning problems. However, this approach significantly increases computational costs and inference time, as the model must generate and evaluate multiple candidate solutions to identify a viable reasoning path. To address this, we propose an effective approach that integrates search capabilities directly into the model by fine-tuning it on unpaired successful (learning) and failed reasoning paths (forgetting) derived from diverse search methods. A key challenge we identify is that naive fine-tuning can degrade the model’s search capability; we show this can be mitigated with a smaller learning rate. Extensive experiments on the challenging Game-of-24 and Countdown arithmetic puzzles show that, replacing CoT-generated data with search-generated data for offline fine-tuning improves success rates by around 23% over inference-time search baselines, while reducing inference time by 180$\times$. On top of this, our learning and forgetting objective consistently outperforms both supervised fine-tuning and preference-based methods.

PDF Details

ICML Conference 2025 Conference Paper

Protein Structure Tokenization: Benchmarking and New Recipe

Xinyu Yuan
Zichen Wang 0002
Marcus D. Collins
Huzefa Rangwala

Recent years have witnessed a surge in the development of protein structural tokenization methods, which chunk protein 3D structures into discrete or continuous representations. Structure tokenization enables the direct application of powerful techniques like language modeling for protein structures, and large multimodal models to integrate structures with protein sequences and functional texts. Despite the progress, the capabilities and limitations of these methods remain poorly understood due to the lack of a unified evaluation framework. We first introduce StructTokenBench, a framework that comprehensively evaluates the quality and efficiency of structure tokenizers, focusing on fine-grained local substructures rather than global structures, as typical in existing benchmarks. Our evaluations reveal that no single model dominates all benchmarking perspectives. Observations of codebook under-utilization led us to develop AminoAseed, a simple yet effective strategy that enhances codebook gradient updates and optimally balances codebook size and dimension for improved tokenizer utilization and quality. Compared to the leading model ESM3, our method achieves an average of 6. 31% performance improvement across 24 supervised tasks, with sensitivity and utilization rates increased by 12. 83% and 124. 03%, respectively. Source code and model weights are available at https: //github. com/KatarinaYuan/StructTokenBench.

Details

ICLR Conference 2025 Conference Paper

Pushing the Limits of All-Atom Geometric Graph Neural Networks: Pre-Training, Scaling, and Zero-Shot Transfer

Zihan Pengmei
Zhengyuan Shen
Zichen Wang 0002
Marcus D. Collins
Huzefa Rangwala

The ability to construct transferable descriptors for molecular and biological systems has broad applications in drug discovery, molecular dynamics, and protein analysis. Geometric graph neural networks (Geom-GNNs) utilizing all-atom information have revolutionized atomistic simulations by enabling the prediction of interatomic potentials and molecular properties. Despite these advances, the application of all-atom Geom-GNNs in protein modeling remains limited due to computational constraints. In this work, we first demonstrate the potential of pre-trained Geom-GNNs as zero-shot transfer learners, effectively modeling protein systems with all-atom granularity. Through extensive experimentation to evaluate their expressive power, we characterize the scaling behaviors of Geom-GNNs across self-supervised, supervised, and unsupervised setups. Interestingly, we find that Geom-GNNs deviate from conventional power-law scaling observed in other domains, with no predictable scaling principles for molecular representation learning. Furthermore, we show how pre-trained graph embeddings can be directly used for analysis and synergize with other architectures to enhance expressive power for protein modeling.

Details

ICLR Conference 2024 Conference Paper

BioBridge: Bridging Biomedical Foundation Models via Knowledge Graphs

Zifeng Wang 0008
Zichen Wang 0002
Balasubramaniam Srinivasan
Vassilis N. Ioannidis
Huzefa Rangwala
Rishita Anubhai

Foundation models (FMs) learn from large volumes of unlabeled data to demonstrate superior performance across a wide range of tasks. However, FMs developed for biomedical domains have largely remained unimodal, i.e., independently trained and used for tasks on protein sequences alone, small molecule structures alone, or clinical data alone. To overcome this limitation, we present BioBridge, a parameter-efficient learning framework, to bridge independently trained unimodal FMs to establish multimodal behavior. BioBridge achieves it by utilizing Knowledge Graphs (KG) to learn transformations between one unimodal FM and another without fine-tuning any underlying unimodal FMs. Our results demonstrate that BioBridge can beat the best baseline KG embedding methods (on average by ~ 76.3%) in cross-modal retrieval tasks. We also identify BioBridge demonstrates out-of-domain generalization ability by extrapolating to unseen modalities or relations. Additionally, we also show that BioBridge presents itself as a general-purpose retriever that can aid biomedical multimodal question answering as well as enhance the guided generation of novel drugs. Code is at https://github.com/RyanWangZf/BioBridge.

Details

ICLR Conference 2024 Conference Paper

Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space

Hengrui Zhang
Jiani Zhang 0003
Zhengyuan Shen
Balasubramaniam Srinivasan
Xiao Qin 0003
Christos Faloutsos
Huzefa Rangwala
George Karypis

Recent advances in tabular data generation have greatly enhanced synthetic data quality. However, extending diffusion models to tabular data is challenging due to the intricately varied distributions and a blend of data types of tabular data. This paper introduces TabSyn, a methodology that synthesizes tabular data by leveraging a diffusion model within a variational autoencoder (VAE) crafted latent space. The key advantages of the proposed Tabsyn include (1) Generality: the ability to handle a broad spectrum of data types by converting them into a single unified space and explicitly capturing inter-column relations; (2) Quality: optimizing the distribution of latent embeddings to enhance the subsequent training of diffusion models, which helps generate high-quality synthetic data; (3) Speed: much fewer number of reverse steps and faster synthesis speed than existing diffusion-based methods. Extensive experiments on six datasets with five metrics demonstrate that Tabsyn outperforms existing methods. Specifically, it reduces the error rates by 86% and 67% for column-wise distribution and pair-wise column correlation estimations compared with the most competitive baselines. The code has been made available at https://github.com/amazon-science/tabsyn.

Details

ICLR Conference 2024 Conference Paper

OpenTab: Advancing Large Language Models as Open-domain Table Reasoners

Kezhi Kong
Jiani Zhang 0003
Zhengyuan Shen
Balasubramaniam Srinivasan
Chuan Lei
Christos Faloutsos
Huzefa Rangwala
George Karypis

Large Language Models (LLMs) trained on large volumes of data excel at various natural language tasks, but they cannot handle tasks requiring knowledge that has not been trained on previously. One solution is to use a retriever that fetches relevant information to expand LLM's knowledge scope. However, existing textual-oriented retrieval-based LLMs are not ideal on structured table data due to diversified data modalities and large table sizes. In this work, we propose OpenTab, an open-domain table reasoning framework powered by LLMs. Overall, OpenTab leverages table retriever to fetch relevant tables and then generates SQL programs to parse the retrieved tables efficiently. Utilizing the intermediate data derived from the SQL executions, it conducts grounded inference to produce accurate response. Extensive experimental evaluation shows that OpenTab significantly outperforms baselines in both open- and closed-domain settings, achieving up to 21.5% higher accuracy. We further run ablation studies to validate the efficacy of our proposed designs of the system.

Details

AAMAS Conference 2023 Conference Paper

Counterfactually Fair Dynamic Assignment: A Case Study on Policing

Tasfia Mashiat
Xavier Gitiaux
Huzefa Rangwala
Sanmay Das

Resource assignment algorithms for decision-making in dynamic environments have been shown to sometimes lead to negative impacts on individuals from minority populations. We propose a framework for algorithmic assignment of scarce resources in a dynamic setting that seeks to minimize concerns around unfairness and the potential for runaway feedback loops that create injustices. Our model estimates an underlying true latent confounder in a biased dataset, and makes allocation decisions based on a notion of fair intervention. We present evidence for the plausibility of our model by analyzing a novel dataset obtained from the City of Chicago through FOIA requests, and plan to release this dataset along with a visualization tool for use by various stakeholders. We also show that, in a simulated environment, our counterfactually fair policy can allocate limited resources near optimally, and better than baseline alternatives.

PDF

IJCAI Conference 2022 Conference Paper

SoFaiR: Single Shot Fair Representation Learning

Xavier Gitiaux
Huzefa Rangwala

To avoid discriminatory uses of their data, organizations can learn to map them into a representation that filters out information related to sensitive attributes. However, all existing methods in fair representation learning generate a fairness-information trade-off. To achieve different points on the fairness-information plane, one must train different models. In this paper, we first demonstrate that fairness-information trade-offs are fully characterized by rate-distortion trade-offs. Then, we use this key result and propose SoFaiR, a single shot fair representation learning method that generates with one trained model many points on the fairness-information plane. Besides its computational saving, our single-shot approach is, to the extent of our knowledge, the first fair representation learning method that explains what information is affected by changes in the fairness / distortion properties of the representation. Empirically, we find on three datasets that SoFaiR achieves similar fairness information trade-offs as its multi-shot counterparts.

PDF Details DOI

AAAI Conference 2021 Conference Paper

Fair Representations by Compression

Xavier Gitiaux
Huzefa Rangwala

Organizations that collect and sell data face increasing scrutiny for the discriminatory use of data. We propose a novel unsupervised approach to transform data into a compressed binary representation independent of sensitive attributes. We show that in an information bottleneck framework, a parsimonious representation should filter out information related to sensitive attributes if they are provided directly to the decoder. Empirical results show that the proposed method, FBC, achieves state-of-the-art accuracyfairness trade-off. Explicit control of the entropy of the representation bit stream allows the user to move smoothly and simultaneously along both rate-distortion and rate-fairness curves.

PDF Details

AAAI Conference 2020 Short Paper

American Sign Language Recognition Using an FMCW Wireless Sensor (Student Abstract)

Yuanqi Du
Nguyen Dang
Riley Wilkerson
Parth Pathak
Huzefa Rangwala
Jana Kosecka

In today’s digital world, rapid technological advancements continue to lessen the burden of tasks for individuals. Among these tasks is communication across perceived language barriers. Indeed, increased attention has been drawn to American Sign Language (ASL) recognition in recent years. Camerabased and motion detection-based methods have been researched extensively; however, there remains a divide in communication between ASL users and non-users. Therefore, this research team proposes the use of a novel wireless sensor (Frequency-Modulated Continuous-Wave Radar) to help bridge the gap in communication. In short, this device sends out signals that detect the user’s body positioning in space. These signals then reﬂect off the body and back to the sensor, developing thousands of cloud points per second, indicating where the body is positioned in space. These cloud points can then be examined for movement over multiple consecutive time frames using a cell division algorithm, ultimately showing how the body moves through space as it completes a single gesture or sentence. At the end of the project, 95% accuracy was achieved in one-object prediction as well as 80% accuracy on cross-object prediction with 30% other objects’ data introduced on 19 commonly used gestures. There are 30 samples for each gesture per person from three persons.

PDF Details

IJCAI Conference 2019 Conference Paper

mdfa: Multi-Differential Fairness Auditor for Black Box Classifiers

Xavier Gitiaux
Huzefa Rangwala

Machine learning algorithms are increasingly involved in sensitive decision-making processes with adversarial implications on individuals. This paper presents a new tool, mdfa that identifies the characteristics of the victims of a classifier's discrimination. We measure discrimination as a violation of multi-differential fairness. Multi-differential fairness is a guarantee that a black box classifier's outcomes do not leak information on the sensitive attributes of a small group of individuals. We reduce the problem of identifying worst-case violations to matching distributions and predicting where sensitive attributes and classifier's outcomes coincide. We apply mdfa to a recidivism risk assessment classifier widely used in the United States and demonstrate that for individuals with little criminal history, identified African-Americans are three-times more likely to be considered at high risk of violent recidivism than similar non-African-Americans.

PDF Details

IJCAI Conference 2013 Conference Paper

Protein Function Prediction by Integrating Multiple Kernels

Guoxian Yu
Huzefa Rangwala
Carlotta Domeniconi
Guoji Zhang
Zili Zhang

Determining protein function constitutes an exercise in integrating information derived from several heterogeneous high-throughput experiments. To utilize the information spread across multiple sources in a combined fashion, these data sources are transformed into kernels. Several protein function prediction methods follow a two-phased approach: they first optimize the weights on individual kernels to produce a composite kernel, and then train a classifier on the composite kernel. As such, these methods result in an optimal composite kernel, but not necessarily in an optimal classifier. On the other hand, some methods optimize the loss of binary classifiers, and learn weights for the different kernels iteratively. A protein has multiple functions, and each function can be viewed as a label. These methods solve the problem of optimizing weights on the input kernels for each of the labels. This is computationally expensive and ignores inter-label correlations. In this paper, we propose a method called Protein Function Prediction by Integrating Multiple Kernels (ProMK). ProMK iteratively optimizes the phases of learning optimal weights and reducing the empirical loss of a multi-label classifier for each of the labels simultaneously, using a combined objective function. ProMK can assign larger weights to smooth kernels and downgrade the weights on noisy kernels. We evaluate the ability of ProMK to predict the function of proteins using several standard benchmarks. We show that our approach performs better than previously proposed protein function prediction approaches that integrate data from multiple networks, and multi-label multiple kernel learning methods.

PDF Details DOI

IJCAI Conference 2011 Conference Paper

A Hidden Markov Model Variant for Sequence Classification

Sam Blasiak
Huzefa Rangwala

Sequence classification is central to many practical problems within machine learning. Distances metrics between arbitrary pairs of sequences can be hard to define because sequences can vary in length and the information contained in the order of sequence elements is lost when standard metrics such as Euclidean distance are applied. We present a scheme that employs a Hidden Markov Model variant to produce a set of fixed-length description vectors from a set of sequences. We then define three inference algorithms, a Baum-Welch variant, a Gibbs Sampling algorithm, and a variational algorithm, to infer model parameters. Finally, we show experimentally that the fixed length representation produced by these inference methods is useful for classifying sequences of amino acids into structural classes

PDF Details DOI

IS Journal 2010 Journal Article

Defining a Coparticipation Network Using Comments on Digg

Huzefa Rangwala
Salman Jamali

The past decade has seen a massive rise in Web services and applications that let users create, collaborate, and share various forms of data including articles (blogs), pictures (Flickr), video (YouTube), and status updates (Twitter). Social bookmarking Web sites such as Delicious. com, Slashdot. org, and Digg. com let users submit links to Web content they find interesting along with a short description. Users in these online communities can comment on the posted content (initiating discussions) and rate the articles they find interesting. Thus, social bookmarking sites serve as data aggregators, Web-based discussion forums, and an online collaborative filtering system that can collectively determine popular online content.

Details DOI