Zhiqi Huang

photo @ Yellowstone

Hi, I am an Applied Researcher at Capital One. I completed my Ph.D. under the mentorship of my advisor, Prof. James Allan, at the Center for Intelligent Information Retrieval (CIIR), Manning College of Information and Computer Sciences, University of Massachusetts Amherst. My research endeavors lie at the intersection of information retrieval, natural language processing, and machine learning (IR/NLP/ML).

I received my master’s in Statistics from the University of Maryland, College Park, and my bachelor’s in Applied Math from Sun Yat-sen University.

news

Oct 01, 2026	Our paper, Memory Busted: Activation-Based Provenance Catches LLMs Cheating on Their Context, has been accepted to the Trustworthy AI for Good (AI4GOOD) Workshop at NeurIPS 2026.
Aug 26, 2026	Our paper, VerTox: Verifiable Reward-Guided Corpus Poisoning Against Neural Ranking Models, has been accepted to the main conference (oral presentation) at EMNLP 2026.
Apr 05, 2026	I’m excited to co-organize MeLLM 2026: The 1st Workshop on Multilinguality in the Era of Large Language Models, to be held at ACL 2026 in San Diego. We welcome paper submissions on all topics related to multilingual LLMs.
Mar 01, 2026	Our paper, Uncertainty as Feature Gaps: Epistemic Uncertainty Quantification of LLMs in Contextual Question-Answering, has been accepted at the ICLR 2026.
Feb 20, 2026	Our survey, A Survey of Model Architectures in Information Retrieval, has been accepted by Transactions on Machine Learning Research (TMLR).

selected publications

EMNLP

VerTox: Verifiable Reward-Guided Corpus Poisoning Against Neural Ranking Models

Zhiqi Huang, Vivek Datla, Zhichao Xu, Puxuan Yu, Vivek Srikumar, and Alfy Samuel

In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026

Abs PDF

Neural ranking models have become core components of modern information retrieval systems and important building blocks of AI systems such as retrieval-augmented generation (RAG) pipelines. However, their robustness remains insufficiently understood in the presence of large language models (LLMs), which can generate fluent and deceptive content at scale. This work investigates the vulnerability of neural ranking models to corpus poisoning attacks, in which an adversary injects a small number of maliciously crafted documents into the corpus to distort ranking behavior. We propose VerTox, the first framework to formulate corpus poisoning as a verifiable reward-guided reinforcement learning (RLVR) problem. By explicitly coupling ranking distortion with factual corruption through specialized reward shaping, we fine-tune compact LLMs into adversarial generators. Experiments demonstrate that our method achieves near-perfect attack success rates, producing adversarial documents that frequently rank higher than target documents across major neural ranking architectures, as well as a proprietary commercial embedding model. The generated adversarial documents are fluent and exhibit low perplexity, making them difficult to detect. Furthermore, by explicitly encouraging factual corruption, our adversarial documents significantly degrade the performance of a downstream RAG application.
ICLR

Uncertainty as Feature Gaps: Epistemic Uncertainty Quantification of LLMs in Contextual Question-Answering

Yavuz Faruk Bakman, Sungmin Kang, Zhiqi Huang, Duygu Nur Yaldiz, Catarina G Belém, Chenyang Zhu, Anoop Kumar, Alfy Samuel, Daben Liu, Salman Avestimehr, and Sai Praneeth Karimireddy

In The Fourteenth International Conference on Learning Representations, 2026

PDF
EMNLP

Language Concept Erasure for Language-invariant Dense Retrieval

Zhiqi Huang, Puxuan Yu, Shauli Ravfogel, and James Allan

In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, 2024

Abs PDF

Multilingual models aim for language-invariant representations but still prominently encode language identity. This, along with the scarcity of high-quality parallel retrieval data, limits their performance in retrieval. We introduce LANCER, a multi-task learning framework that improves language-invariant dense retrieval by reducing language-specific signals in the embedding space. Leveraging the notion of linear concept erasure, we design a loss function that penalizes cross-correlation between representations and their language labels. LANCER leverages only English retrieval data and general multilingual corpora, training models to focus on language-invariant retrieval by semantic similarity without necessitating a vast parallel corpus. Experimental results on various datasets show our method consistently improves over baselines, with extensive analyses demonstrating greater language agnosticism.