Zhichao Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhichao Xu has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
Vivek Srikumar
Kahlert School of Computing University Of Utah
3 shared publicationsBei Wang
Kahlert School of Computing University Of Utah
2 shared publications- Chia-Yuan Chang
Texas A&M University
2 shared publications Daniel Cohen
Dataminr Inc
2 shared publications- Mingzhi Hu
Worcester Polytechnic Institute
2 shared publications - Na Zou
University of Houston
2 shared publications
Latest Publications
- Open Access
LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
While dense retrieval models have become the standard for state-of-the-art information retrieval, their deployment is often constrained by high memory requirements and reliance on GPU accelerators for vector similarity search. Learned sparse retrieval offers a compelling alternative by enabling efficient search via inverted indices, yet it has historically received less attention than dense approaches. In this report, we introduce LACONIC, a family of learned sparse retrievers based on the Llama-3 architecture (1B, 3B, and 8B). We propose a streamlined two-phase training curriculum consisting of (1) weakly supervised pre-finetuning to adapt causal LLMs for bidirectional contextualization and (2) high-signal finetuning using curated hard negatives. Our results demonstrate that LACONIC effectively bridges the performance gap with dense models: the 8B variant achieves a state-of-the-art 60.2 nDCG on the MTEB Retrieval benchmark, ranking 15th on the leaderboard as of January 1, 2026, while utilizing 71\% less index memory than an equivalent dense model. By delivering high retrieval effectiveness on commodity CPU hardware with a fraction of the compute budget required by competing models, LACONIC provides a scalable and efficient solution for real-world search applications.
Open AccessLACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
While dense retrieval models have become the standard for state-of-the-art information retrieval, their deployment is often constrained by high memory requirements and reliance on GPU accelerators for vector similarity search. Learned sparse retrieval offers a compelling alternative by enabling efficient search via inverted indices, yet it has historically received less attention than dense approaches. In this report, we introduce LACONIC, a family of learned sparse retrievers based on the Llama-3 architecture (1B, 3B, and 8B). We propose a streamlined two-phase training curriculum consisting of (1) weakly supervised pre-finetuning to adapt causal LLMs for bidirectional contextualization and (2) high-signal finetuning using curated hard negatives. Our results demonstrate that LACONIC effectively bridges the performance gap with dense models: the 8B variant achieves a state-of-the-art 60.2 nDCG on the MTEB Retrieval benchmark, ranking 15th on the leaderboard as of January 1, 2026, while utilizing 71\% less index memory than an equivalent dense model. By delivering high retrieval effectiveness on commodity CPU hardware with a fraction of the compute budget required by competing models, LACONIC provides a scalable and efficient solution for real-world search applications.
- Open Access
RECON: Reasoning with Condensation for Efficient Retrieval-Augmented Generation
Retrieval-augmented generation (RAG) systems trained using reinforcement learning (RL) with reasoning are hampered by inefficient context management, where long, noisy retrieved documents increase costs and degrade performance. We introduce RECON (REasoning with CONdensation), a framework that integrates an explicit summarization module to compress evidence within the reasoning loop. Our summarizer is trained via a two-stage process: relevance pretraining on QA datasets, followed by multi-aspect distillation from proprietary LLMs to ensure factuality and clarity. Integrated into the Search-R1 pipeline, RECON reduces total context length by 35\%, leading to improved training speed and inference latency, while simultaneously improving RAG performance on downstream QA benchmarks. Notably, it boosts the average EM score of the 3B model by 14.5\% and the 7B model by 3.0\%, showing particular strength in multi-hop QA. RECON demonstrates that learned context compression is essential for building practical, scalable, and performant RAG systems. Our code implementation is made available at https://github.com/allfornancy/RECON.
Open AccessRECON: Reasoning with Condensation for Efficient Retrieval-Augmented Generation
Retrieval-augmented generation (RAG) systems trained using reinforcement learning (RL) with reasoning are hampered by inefficient context management, where long, noisy retrieved documents increase costs and degrade performance. We introduce RECON (REasoning with CONdensation), a framework that integrates an explicit summarization module to compress evidence within the reasoning loop. Our summarizer is trained via a two-stage process: relevance pretraining on QA datasets, followed by multi-aspect distillation from proprietary LLMs to ensure factuality and clarity. Integrated into the Search-R1 pipeline, RECON reduces total context length by 35\%, leading to improved training speed and inference latency, while simultaneously improving RAG performance on downstream QA benchmarks. Notably, it boosts the average EM score of the 3B model by 14.5\% and the 7B model by 3.0\%, showing particular strength in multi-hop QA. RECON demonstrates that learned context compression is essential for building practical, scalable, and performant RAG systems. Our code implementation is made available at https://github.com/allfornancy/RECON.
- Top Cited PapersOpen Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 2607-2622
MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
Chia-Yuan Chang, Zhimeng Jiang, Vineeth Rakesh, Menghai Pan, Chin-Chia Michael Yeh, Guanchu Wang, Mingzhi Hu, Zhichao Xu, Yan Zheng, Mahashweta Das, Na Zou. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 2607-2622
Top Cited PapersOpen AccessMAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
Chia-Yuan Chang, Zhimeng Jiang, Vineeth Rakesh, Menghai Pan, Chin-Chia Michael Yeh, Guanchu Wang, Mingzhi Hu, Zhichao Xu, Yan Zheng, Mahashweta Das, Na Zou. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- Open Access
MAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
Large Language Models (LLMs) are becoming essential tools for various natural language processing tasks but often suffer from generating outdated or incorrect information. Retrieval-Augmented Generation (RAG) addresses this issue by incorporating external, real-time information retrieval to ground LLM responses. However, the existing RAG systems frequently struggle with the quality of retrieval documents, as irrelevant or noisy documents degrade performance, increase computational overhead, and undermine response reliability. To tackle this problem, we propose Multi-Agent Filtering Retrieval-Augmented Generation (MAIN-RAG), a training-free RAG framework that leverages multiple LLM agents to collaboratively filter and score retrieved documents. Specifically, MAIN-RAG introduces an adaptive filtering mechanism that dynamically adjusts the relevance filtering threshold based on score distributions, effectively minimizing noise while maintaining high recall of relevant documents. The proposed approach leverages inter-agent consensus to ensure robust document selection without requiring additional training data or fine-tuning. Experimental results across four QA benchmarks demonstrate that MAIN-RAG consistently outperforms traditional RAG approaches, achieving a 2-11% improvement in answer accuracy while reducing the number of irrelevant retrieved documents. Quantitative analysis further reveals that our approach achieves superior response consistency and answer accuracy over baseline methods, offering a competitive and practical alternative to training-based solutions.
Open AccessMAIN-RAG: Multi-Agent Filtering Retrieval-Augmented Generation
Large Language Models (LLMs) are becoming essential tools for various natural language processing tasks but often suffer from generating outdated or incorrect information. Retrieval-Augmented Generation (RAG) addresses this issue by incorporating external, real-time information retrieval to ground LLM responses. However, the existing RAG systems frequently struggle with the quality of retrieval documents, as irrelevant or noisy documents degrade performance, increase computational overhead, and undermine response reliability. To tackle this problem, we propose Multi-Agent Filtering Retrieval-Augmented Generation (MAIN-RAG), a training-free RAG framework that leverages multiple LLM agents to collaboratively filter and score retrieved documents. Specifically, MAIN-RAG introduces an adaptive filtering mechanism that dynamically adjusts the relevance filtering threshold based on score distributions, effectively minimizing noise while maintaining high recall of relevant documents. The proposed approach leverages inter-agent consensus to ensure robust document selection without requiring additional training data or fine-tuning. Experimental results across four QA benchmarks demonstrate that MAIN-RAG consistently outperforms traditional RAG approaches, achieving a 2-11% improvement in answer accuracy while reducing the number of irrelevant retrieved documents. Quantitative analysis further reveals that our approach achieves superior response consistency and answer accuracy over baseline methods, offering a competitive and practical alternative to training-based solutions.
- Open Access
In-Context Example Ordering Guided by Label Distributions
By allowing models to predict without task-specific training, in-context learning (ICL) with pretrained LLMs has enormous potential in NLP. However, a number of problems persist in ICL. In particular, its performance is sensitive to the choice and order of in-context examples. Given the same set of in-context examples with different orderings, model performance may vary between near random to near state-of-the-art. In this work, we formulate in-context example ordering as an optimization problem. We examine three problem settings that differ in the assumptions they make about what is known about the task. Inspired by the idea of learning from label proportions, we propose two principles for in-context example ordering guided by model's probability predictions. We apply our proposed principles to thirteen text classification datasets and nine different autoregressive LLMs with 700M to 13B parameters. We demonstrate that our approach outperforms the baselines by improving the classification accuracy, reducing model miscalibration, and also by selecting better in-context examples.
Open AccessIn-Context Example Ordering Guided by Label Distributions
By allowing models to predict without task-specific training, in-context learning (ICL) with pretrained LLMs has enormous potential in NLP. However, a number of problems persist in ICL. In particular, its performance is sensitive to the choice and order of in-context examples. Given the same set of in-context examples with different orderings, model performance may vary between near random to near state-of-the-art. In this work, we formulate in-context example ordering as an optimization problem. We examine three problem settings that differ in the assumptions they make about what is known about the task. Inspired by the idea of learning from label proportions, we propose two principles for in-context example ordering guided by model's probability predictions. We apply our proposed principles to thirteen text classification datasets and nine different autoregressive LLMs with 700M to 13B parameters. We demonstrate that our approach outperforms the baselines by improving the classification accuracy, reducing model miscalibration, and also by selecting better in-context examples.
- Open Access
An In-depth Investigation of User Response Simulation for Conversational Search
Conversational search has seen increased recent attention in both the IR and NLP communities. It seeks to clarify and solve users' search needs through multi-turn natural language interactions. However, most existing systems are trained and demonstrated with recorded or artificial conversation logs. Eventually, conversational search systems should be trained, evaluated, and deployed in an open-ended setting with unseen conversation trajectories. A key challenge is that training and evaluating such systems both require a human-in-the-loop, which is expensive and does not scale. One strategy is to simulate users, thereby reducing the scaling costs. However, current user simulators are either limited to only responding to yes-no questions from the conversational search system or unable to produce high-quality responses in general. In this paper, we show that existing user simulation systems could be significantly improved by a smaller finetuned natural language generation model. However, rather than merely reporting it as the new state-of-the-art, we consider it a strong baseline and present an in-depth investigation of simulating user response for conversational search. Our goal is to supplement existing work with an insightful hand-analysis of unsolved challenges by the baseline and propose our solutions. The challenges we identified include (1) a blind spot that is difficult to learn, and (2) a specific type of misevaluation in the standard setup. We propose a new generation system to effectively cover the training blind spot and suggest a new evaluation setup to avoid misevaluation. Our proposed system leads to significant improvements over existing systems and large language models such as GPT-4. Additionally, our analysis provides insights into the nature of user simulation to facilitate future work.
Open AccessAn In-depth Investigation of User Response Simulation for Conversational Search
Conversational search has seen increased recent attention in both the IR and NLP communities. It seeks to clarify and solve users' search needs through multi-turn natural language interactions. However, most existing systems are trained and demonstrated with recorded or artificial conversation logs. Eventually, conversational search systems should be trained, evaluated, and deployed in an open-ended setting with unseen conversation trajectories. A key challenge is that training and evaluating such systems both require a human-in-the-loop, which is expensive and does not scale. One strategy is to simulate users, thereby reducing the scaling costs. However, current user simulators are either limited to only responding to yes-no questions from the conversational search system or unable to produce high-quality responses in general. In this paper, we show that existing user simulation systems could be significantly improved by a smaller finetuned natural language generation model. However, rather than merely reporting it as the new state-of-the-art, we consider it a strong baseline and present an in-depth investigation of simulating user response for conversational search. Our goal is to supplement existing work with an insightful hand-analysis of unsolved challenges by the baseline and propose our solutions. The challenges we identified include (1) a blind spot that is difficult to learn, and (2) a specific type of misevaluation in the standard setup. We propose a new generation system to effectively cover the training blind spot and suggest a new evaluation setup to avoid misevaluation. Our proposed system leads to significant improvements over existing systems and large language models such as GPT-4. Additionally, our analysis provides insights into the nature of user simulation to facilitate future work.
- Top Cited PapersOpen Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2066-2087
Multi-dimensional Evaluation of Empathetic Dialogue Responses
Zhichao Xu, Jiepu Jiang. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2066-2087
Top Cited PapersOpen AccessMulti-dimensional Evaluation of Empathetic Dialogue Responses
Zhichao Xu, Jiepu Jiang. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 15359-15396
Beyond Perplexity: Multi-dimensional Safety Evaluation of LLM Compression
Zhichao Xu, Ashim Gupta, Tao Li, Oliver Bentham, Vivek Srikumar. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 15359-15396
Open AccessBeyond Perplexity: Multi-dimensional Safety Evaluation of LLM Compression
Zhichao Xu, Ashim Gupta, Tao Li, Oliver Bentham, Vivek Srikumar. Findings of the Association for Computational Linguistics: EMNLP 2024. 2024.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2623-2640
In-Context Example Ordering Guided by Label Distributions
Zhichao Xu, Daniel Cohen, Bei Wang, Vivek Srikumar. Findings of the Association for Computational Linguistics: NAACL 2024. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 2623-2640
Open AccessIn-Context Example Ordering Guided by Label Distributions
Zhichao Xu, Daniel Cohen, Bei Wang, Vivek Srikumar. Findings of the Association for Computational Linguistics: NAACL 2024. 2024.