Author name cluster

Zhicheng Dou

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.

19 papers

2 author rows

AAAI Conference 2026 System Paper

Evaluating the Factuality of Large Language Models Using Multiple Plug-and-Play Fact Sources

Zhaoheng Huang
Yutao Zhu
Jirong Wen
Zhicheng Dou

Large language models (LLMs) often produce factually inaccurate content, or hallucinations, which undermines their reliability. Existing factuality evaluation systems usually rely on a single predefined fact source, making them task-specific and hard to extend. We present UFO, a unified framework for factuality evaluation that supports multiple plug-and-play fact sources. UFO integrates human-written evidence, web search results, and LLM knowledge within a single evaluation pipeline, and allows users to flexibly select, reorder, and even define customized sources. The system is accessible through both a Python interface and a web-based demo, offering interactive claim-level verification and visualization. Experiments show that UFO system achieves moderate consistency with human annotations. Overall, UFO serves as a transparent and extensible platform for benchmarking fact sources, comparing LLMs, and enabling real-world fact-checking applications across diverse domains.

PDF Details DOI

AAAI Conference 2026 Conference Paper

HierSearch: A Hierarchical Enterprise Deep Search Framework Integrating Local and Web Searches

Jiejun Tan
Zhicheng Dou
Yan Yu
Jiehan Cheng
Lifeng Liu
Jian Xie
Jirong Wen

Recently, large reasoning models have demonstrated strong mathematical and coding abilities, and deep search leverages their reasoning capabilities in challenging information retrieval tasks. Existing deep search works are generally limited to a single knowledge source, either local or the Web. However, enterprises often require private deep search systems that can leverage search tools over both local and the Web corpus. Simply training an agent equipped with multiple search tools using flat reinforcement learning (RL) is a straightforward idea, but it has problems such as low training data efficiency and poor mastery of complex tools. To address the above issue, we propose a hierarchical agentic deep search framework, HierSearch, trained with hierarchical RL. At the low level, a local deep search agent and a Web deep search agent are trained to retrieve evidence from their corresponding domains. At the high level, a planner agent coordinates low-level agents and provides the final answer. Moreover, to prevent direct answer copying and error propagation, we design a knowledge refiner that filters out hallucinations and irrelevant evidence returned by low-level agents. Experiments show that HierSearch achieves better performance compared to flat RL, and outperforms various deep search and multi-source retrieval-augmented generation baselines in six benchmarks across general, finance, and medical domains.

PDF Details DOI

ICLR Conference 2025 Conference Paper

A Non-Contrastive Learning Framework for Sequential Recommendation with Preference-Preserving Profile Generation

Huimin Zeng
Xiaojie Wang 0003
Anoop Jain
Zhicheng Dou
Dong Wang

Contrastive Learning (CL) proves to be effective for learning generalizable user representations in Sequential Recommendation (SR), but it suffers from high computational costs due to its reliance on negative samples. To overcome this limitation, we propose the first Non-Contrastive Learning (NCL) framework for SR, which eliminates computational overhead of identifying and generating negative samples. However, without negative samples, it is challenging to learn uniform representations from only positive samples, which is prone to representation collapse. Furthermore, the alignment of the learned representations may be substantially compromised because existing ad-hoc augmentations can produce positive samples that have inconsistent user preferences. To tackle these challenges, we design a novel preference-preserving profile generation method to produce high-quality positive samples for non-contrastive training. Inspired by differential privacy, our approach creates augmented user profiles that exhibit high diversity while provably retaining consistent user preferences. With larger diversity and consistency of the positive samples, our NCL framework significantly enhances the alignment and uniformity of the learned representations, which contributes to better generalization. The experimental results on various benchmark datasets and model architectures demonstrate the effectiveness of the proposed method. Finally, our investigations reveal that both uniformity and alignment play a vital role in improving generalization for SR. Interestingly, in our data-sparse setting, alignment is usually more important than uniformity.

Details

NeurIPS Conference 2025 Conference Paper

Chain-of-Retrieval Augmented Generation

Liang Wang
Haonan Chen
Nan Yang
Xiaolong Huang
Zhicheng Dou
Furu Wei

This paper introduces an approach for training o1-like RAG models that retrieve and reason over relevant information step by step before generating the final answer. Conventional RAG methods usually perform a single retrieval step before the generation process, which limits their effectiveness in addressing complex queries due to imperfect retrieval results. In contrast, our proposed method, CoRAG (Chain-of-Retrieval Augmented Generation), allows the model to dynamically reformulate the query based on the evolving state. To train CoRAG effectively, we utilize rejection sampling to automatically generate intermediate retrieval chains, thereby augmenting existing RAG datasets that only provide the correct final answer. At test time, we propose various decoding strategies to scale the model's test-time compute by controlling the length and number of sampled retrieval chains. Experimental results across multiple benchmarks validate the efficacy of CoRAG, particularly in multi-hop question answering tasks, where we observe more than $10$ points improvement in EM score compared to strong baselines. On the KILT benchmark, CoRAG establishes a new state-of-the-art performance across a diverse range of knowledge-intensive tasks. Furthermore, we offer comprehensive analyses to understand the scaling behavior of CoRAG, laying the groundwork for future research aimed at developing factual and grounded foundation models.

PDF Details

AAAI Conference 2025 Conference Paper

Descriptive and Discriminative Document Identifiers for Generative Retrieval

Jiehan Cheng
Zhicheng Dou
Yutao Zhu
Xiaoxi Li

Generative document retrieval is a novel retrieval framework, which represents documents as identifiers (DocID) and retrieves documents by generating DocIDs. It has the advantage of end-to-end optimization over traditional retrieval methods and has attracted much research interest. Nonetheless, the development of efficient and precise DocIDs for document representation remains a pertinent issue within the field. Existing methods for designing DocIDs tend to consider only the relevance of DocIDs to the corresponding documents, while neglecting the ability of the DocIDs to distinguish the corresponding documents from similar ones, which is crucial for the retrieval task. In this paper, we design learnable descriptive and discriminative document Identifiers (D2-DocID) for Generative Retrieval and propose the paired retrieval model D2Gen. The D2-DocID is semantically similar to the corresponding documents (descriptive) and is able to distinguish similar documents (discriminative) in the corpus, thus enhancing retrieval performance. We use a contrastive learning assisted generative retrieval task to enable the model to understand the document and then complete the generative retrieval. We then design a DocID selection method to select DocIDs based on the retrieval model's understanding of the documents. Our experimental results on the MS MARCO and NQ320k dataset illustrate the effectiveness of the approach.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

HawkBench: Investigating Resilience of RAG Methods on Stratified Information-Seeking Tasks

Hongjin Qian
Zheng Liu
Chao Gao
Yankai Wang
Defu Lian
Zhicheng Dou

In real-world information-seeking scenarios, users have dynamic and diverse needs, requiring RAG systems to demonstrate adaptable resilience. To comprehensively evaluate the resilience of current RAG methods, we introduce HawkBench, a human-labeled, multi-domain benchmark designed to rigorously assess RAG performance across categorized task types. By stratifying tasks based on information-seeking behaviors, HawkBench provides a systematic evaluation of how well RAG systems adapt to diverse user needs. Unlike existing benchmarks, which focus primarily on specific task types (mostly factoid queries) and rely on varying knowledge bases, HawkBench offers: (1) systematic task stratification to cover a broad range of query types, including both factoid and rationale queries, (2) integration of multi-domain corpora across all task types to mitigate corpus bias, and (3) rigorous annotation for high-quality evaluation. HawkBench includes 1, 600 high-quality test samples, evenly distributed across domains and task types. Using this benchmark, we evaluate representative RAG methods, analyzing their performance in terms of answer quality and response latency. Our findings highlight the need for dynamic task strategies that integrate decision-making, query interpretation, and global knowledge understanding to improve RAG generalizability. We believe HawkBench serves as a pivotal benchmark for advancing the resilience of RAG methods and their ability to achieve general-purpose information seeking.

PDF Details

ICLR Conference 2025 Conference Paper

Long Context Compression with Activation Beacon

Peitian Zhang
Zheng Liu 0011
Shitao Xiao
Ninglu Shao
Qiwei Ye
Zhicheng Dou

Long context compression is a critical research problem due to its significance in reducing the high computational and memory costs associated with LLMs. In this paper, we propose Activation Beacon, a plug-in module for transformer-based LLMs that targets effective, efficient, and flexible compression of long contexts. To achieve this, our method introduces the following technical designs. 1) We directly compress the activations (i.e. keys and values at every layer), rather than leveraging soft prompts to relay information (which constitute a major bottleneck to encapsulate the complex information within long contexts). 2) We tailor the compression workflow, where each fine-grained input unit is progressively compressed, enabling high-quality compression and efficient computation during both training and inference. 3) We train the model through compression-based auto-regression, making full use of plain texts and instructional data to optimize the model's compression performance. 4) During training, we randomly sample a compression ratio at each step, teaching the model to support a wide range of compression configurations. Extensive evaluations are conducted on various long-context tasks whose lengths (e.g., 128K) may far exceed the maximum training length (20K), such as document understanding, few-shot learning, and Needle-in-a-Haystack. Whilst existing methods struggle to handle these challenging tasks, Activation Beacon maintains a comparable performance to the uncompressed baseline across various scenarios, achieving a 2x acceleration in inference time and an 8x reduction of memory costs for KV cache.

Details

NeurIPS Conference 2025 Conference Paper

MomentSeeker: A Task-Oriented Benchmark For Long-Video Moment Retrieval

Huaying Yuan
Jian Ni
Zheng Liu
Yueze Wang
Junjie Zhou
Zhengyang Liang
Bo Zhao
Zhao Cao

Accurately locating key moments within long videos is crucial for solving long video understanding (LVU) tasks. However, existing benchmarks are either severely limited in terms of video length and task diversity, or they focus solely on the end-to-end LVU performance, making them inappropriate for evaluating whether key moments can be accurately accessed. To address this challenge, we propose MomentSeeker, a novel benchmark for long-video moment retrieval (LVMR), distinguished by the following features. First, it is created based on long and diverse videos, averaging over 1, 200 seconds in duration, and collected from various domains, e. g. , movie, anomaly, egocentric, and sports. Second, it covers a variety of real-world scenarios in three levels: global-level, event-level, and object-level, covering common tasks like action recognition, object localization, causal reasoning, etc. Third, it incorporates rich forms of queries, including text-only queries, image-conditioned queries, and video-conditioned queries. On top of MomentSeeker, we conduct comprehensive experiments for both generation-based approaches (directly using MLLMs) and retrieval-based approaches (leveraging video retrievers). Our results reveal the significant challenges in long-video moment retrieval in terms of accuracy and efficiency, despite improvements from the latest long-video MLLMs and task-specific fine-tuning. We have publicly released MomentSeeker to facilitate future research in this area.

PDF Details

AAAI Conference 2025 Conference Paper

One Token Can Help! Learning Scalable and Pluggable Virtual Tokens for Retrieval-Augmented Large Language Models

Yutao Zhu
Zhaoheng Huang
Zhicheng Dou
Ji-Rong Wen

Retrieval-augmented generation (RAG) is a promising way to improve large language models (LLMs) for generating more factual, accurate, and up-to-date content. Existing methods either optimize prompts to guide LLMs in leveraging retrieved information or directly fine-tune LLMs to adapt to RAG scenarios. Although fine-tuning can yield better performance, it often compromises the LLMs' general generation capabilities by modifying their parameters. This limitation poses challenges in practical applications, especially when LLMs are already deployed, as parameter adjustments may affect their original functionality. To address this, we propose a novel method that involves learning scalable and pluggable virtual tokens for RAG. By maintaining the LLMs' original parameters and fine-tuning only the embeddings of these pluggable tokens, our approach not only enhances LLMs' performance but also preserves their general generation capabilities. Furthermore, we design several training strategies to improve the scalability, flexibility, and generalizability of our method. Comprehensive experiments across 12 question-answering tasks demonstrate the superiority of our approach.

PDF Details DOI

AAAI Conference 2025 Conference Paper

Toward Verifiable Instruction-Following Alignment for Retrieval Augmented Generation

Guanting Dong
Xiaoshuai Song
Yutao Zhu
Runqi Qiao
Zhicheng Dou
Ji-Rong Wen

Following natural instructions is crucial for the effective application of Retrieval-Augmented Generation (RAG) systems. Despite recent advancements in Large Language Models (LLMs), research on assessing and improving instruction-following (IF) alignment within the RAG domain remains limited. To address this issue, we propose VIF-RAG, an automated, scalable, and verifiable synthetic pipeline for instruction-following alignment in RAG systems. We start by manually crafting a minimal set of atomic instructions ( 100k) through automated processes. To further bridge the gap in instruction-following auto-evaluation for RAG systems, we introduce FollowRAG Benchmark, which includes approximately 3K test samples, covering 22 categories of general instruction constraints and four knowledge-intensive QA datasets. Due to its robust pipeline design, FollowRAG can seamlessly integrate with different RAG benchmarks. Using FollowRAG and eight widely-used IF and foundational abilities benchmarks for LLMs, we demonstrate that VIF-RAG markedly enhances LLM performance across a broad range of general instruction constraints while effectively leveraging its capabilities in RAG scenarios. Further analysis offers practical insights for achieving IF alignment in RAG systems.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

UniGist: Towards General and Hardware-aligned Sequence-level Long Context Compression

Chenlong Deng
Zhisong Zhang
Kelong Mao
Shuaiyi Li
Tianqing Fang
Hongming Zhang
Haitao Mi
Dong Yu

Large language models are increasingly capable of handling long-context inputs, but the memory overhead of KV cache remains a major bottleneck for general-purpose deployment. While many compression strategies have been explored, sequence-level compression is particularly challenging due to its tendency to lose important details. We present UniGist, a gist token-based long context compression framework that removes the need for chunk-wise training, enabling the model to learn how to compress and utilize long-range context during training. To fully exploit the sparsity, we introduce a gist shift trick that transforms the attention layout into a right-aligned block structure and develop a block-table-free sparse attention kernel based on it. UniGist further supports one-pass training and flexible chunk sizes during inference, allowing efficient and adaptive context processing. Experiments across multiple long-context tasks show that UniGist significantly improves compression quality, with especially strong performance in recalling details and long-range dependency modeling.

PDF Details

NeurIPS Conference 2025 Conference Paper

WebThinker: Empowering Large Reasoning Models with Deep Research Capability

Xiaoxi Li
Jiajie Jin
Guanting Dong
Hongjin Qian
Yongkang Wu
Ji-Rong Wen
Yutao Zhu
Zhicheng Dou

Large reasoning models (LRMs), such as OpenAI-o1 and DeepSeek-R1, demonstrate impressive long-horizon reasoning capabilities. However, their reliance on static internal knowledge limits their performance on complex, knowledge-intensive tasks and hinders their ability to produce comprehensive research reports requiring synthesis of diverse web information. To address this, we propose WebThinker, a deep research agent that empowers LRMs to autonomously search the web, navigate among web pages, and draft reports during the reasoning process. WebThinker integrates a Deep Web Explorer module, enabling LRMs to dynamically search, navigate, and extract information from the web when encountering knowledge gaps. It also employs an Autonomous Think-Search-and-Draft strategy, allowing the model to seamlessly interleave reasoning, information gathering, and report writing in real time. To further enhance research tool utilization, we introduce an RL-based training strategy via iterative online Direct Preference Optimization (DPO). Extensive experiments on complex reasoning benchmarks (GPQA, GAIA, WebWalkerQA, HLE) and scientific report generation tasks (Glaive) demonstrate that WebThinker significantly outperforms existing methods and strong proprietary systems. Our approach enhances LRM reliability and applicability in complex scenarios, paving the way for more capable and versatile deep research systems. The code is available at https: //github. com/RUC-NLPIR/WebThinker.

PDF Details

NeurIPS Conference 2024 Conference Paper

Boosting the Potential of Large Language Models with an Intelligent Information Assistant

Yujia Zhou
Zheng Liu
Zhicheng Dou

The emergence of Large Language Models (LLMs) has significantly advanced natural language processing, but these models often generate factually incorrect information, known as "hallucination. " Initial retrieval-augmented generation (RAG) methods like the "Retrieve-Read" framework was inadequate for complex reasoning tasks. Subsequent prompt-based RAG strategies and Supervised Fine-Tuning (SFT) methods improved performance but required frequent retraining and risked altering foundational LLM capabilities. To cope with these challenges, we propose Assistant-based Retrieval-Augmented Generation (AssistRAG), integrating an intelligent information assistant within LLMs. This assistant manages memory and knowledge through tool usage, action execution, memory building, and plan specification. Using a two-phase training approach—Curriculum Assistant Learning and Reinforced Preference Optimization—AssistRAG enhances information retrieval and decision-making. Experiments show AssistRAG significantly outperforms benchmarks, especially benefiting less advanced LLMs, by providing superior reasoning capabilities and accurate responses.

PDF Details DOI

AAAI Conference 2024 Conference Paper

UniGen: A Unified Generative Framework for Retrieval and Question Answering with Large Language Models

Xiaoxi Li
Yujia Zhou
Zhicheng Dou

Generative information retrieval, encompassing two major tasks of Generative Document Retrieval (GDR) and Grounded Answer Generation (GAR), has gained significant attention in natural language processing. Existing methods for GDR and GAR rely on separate retrieval and reader modules, which hinder simultaneous optimization. To overcome this, we present UniGen, a Unified Generative framework for retrieval and question answering that integrates both tasks into a single generative model leveraging the capabilities of large language models. UniGen employs a shared encoder and two distinct decoders for generative retrieval and question answering. To facilitate the learning of both tasks, we introduce connectors, generated by large language models, to bridge the gaps between query inputs and generation targets, as well as between document identifiers and answers. Furthermore, we propose an iterative enhancement strategy that leverages generated answers and retrieved documents to iteratively improve both tasks. Through extensive experiments on the MS MARCO and NQ datasets, we demonstrate the effectiveness of UniGen, showcasing its superior performance in both retrieval and question answering tasks.

PDF Details DOI

TIST Journal 2023 Journal Article

CDSM: C ascaded D eep S emantic M atching on Textual Graphs Leveraging Ad-hoc Neighbor Selection

Jing Yao
Zheng Liu
Junhan Yang
Zhicheng Dou
Xing Xie
Ji-Rong Wen

Deep semantic matching aims at discriminating the relationship between documents based on deep neural networks. In recent years, it becomes increasingly popular to organize documents with a graph structure, then leverage both the intrinsic document features and the extrinsic neighbor features to derive discrimination. Most of the existing works mainly care about how to utilize the presented neighbors, whereas limited effort is made to filter appropriate neighbors. We argue that the neighbor features could be highly noisy and partially useful. Thus, a lack of effective neighbor selection will not only incur a great deal of unnecessary computation cost but also restrict the matching accuracy severely. In this work, we propose a novel framework, C ascaded D eep S emantic M atching ( CDSM ), for accurate and efficient semantic matching on textual graphs. CDSM is highlighted for its two-stage workflow. In the first stage, a lightweight CNN-based ad-hod neighbor selector is deployed to filter useful neighbors for the matching task with a small computation cost. We design both one-step and multi-step selection methods. In the second stage, a high-capacity graph-based matching network is employed to compute fine-grained relevance scores based on the well-selected neighbors. It is worth noting that CDSM is a generic framework which accommodates most of the mainstream graph-based semantic matching networks. The major challenge is how the selector can learn to discriminate the neighbors’ usefulness which has no explicit labels. To cope with this problem, we design a weak-supervision strategy for optimization, where we train the graph-based matching network at first and then the ad-hoc neighbor selector is learned on top of the annotations from the matching network. We conduct extensive experiments with three large-scale datasets, showing that CDSM notably improves the semantic matching accuracy and efficiency thanks to the selection of high-quality neighbors. The source code is released at https://github.com/jingjyyao/CDSM.

Details DOI

AAAI Conference 2023 Conference Paper

Learning from the Wisdom of Crowds: Exploiting Similar Sessions for Session Search

Yuhang Ye
Zhonghua Li
Zhicheng Dou
Yutao Zhu
Changwang Zhang
Shangquan Wu
Zhao Cao

Search engines are essential internet services, enabling users to efficiently find the information they need. Session search employs users’ session logs of queries to solve complex retrieval tasks, in which users search multiple times until interested documents are found. Most existing session search models focus on the contextual information within the current search, ignoring the evidence from historical search sessions. Considering the fact that many ongoing retrieval tasks should have already been carried out by other users with a similar intent, we argue that historical sessions with similar intents can help improve the accuracy of the current search task. We propose a novel Similar Session-enhanced Ranking (SSR) model to improve the session search performance using historical sessions with similar intents. Specifically, the candidate historical sessions are matched by query-level and session-level semantic similarity, and then query-level neighbor behaviors are aggregated by a Query-guided GNN (QGNN) while session-level neighbor behaviors are aggregated using the attention mechanism. Finally, we integrate the refined and aggregated historical neighbor information into the current search session. Experimental results on AOL and Tiangong-ST datasets show that our SSR model significantly outperforms the state-of-the-art models.

PDF Details DOI

AAAI Conference 2021 Conference Paper

Neural Sentence Ordering Based on Constraint Graphs

Yutao Zhu
Kun Zhou
Jian-Yun Nie
Shengchao Liu
Zhicheng Dou

Sentence ordering aims at arranging a list of sentences in the correct order. Based on the observation that sentence order at different distances may rely on different types of information, we devise a new approach based on multi-granular orders between sentences. These orders form multiple constraint graphs, which are then encoded by Graph Isomorphism Networks and fused into sentence representations. Finally, sentence order is determined using the order-enhanced sentence representations. Our experiments on five benchmark datasets show that our method outperforms all existing baselines significantly, achieving a new state-of-the-art performance. The results demonstrate the advantage of considering multiple types of order information and using graph neural networks to integrate sentence content and order information for the task. Our code is available at https: //github. com/DaoD/ ConstraintGraph4NSO.

PDF Details

AAAI Conference 2020 Conference Paper

Leveraging Multi-Token Entities in Document-Level Named Entity Recognition

Anwen Hu
Zhicheng Dou
Jian-Yun Nie
Ji-Rong Wen

Most state-of-the-art named entity recognition systems are designed to process each sentence within a document independently. These systems are easy to confuse entity types when the context information in a sentence is not sufﬁcient enough. To utilize the context information within the whole document, most document-level work let neural networks on their own to learn the relation across sentences, which is not intuitive enough for us humans. In this paper, we divide entities to multi-token entities that contain multiple tokens and single-token entities that are composed of a single token. We propose that the context information of multi-token entities should be more reliable in document-level NER for news articles. We design a fusion attention mechanism which not only learns the semantic relevance between occurrences of the same token, but also focuses more on occurrences belonging to multi-tokens entities. To identify multi-token entities, we design an auxiliary task namely ‘Multi-token Entity Classiﬁcation’ and perform this task simultaneously with document-level NER. This auxiliary task is simpliﬁed from NER and doesn’t require extra annotation. Experimental results on the CoNLL-2003 dataset and OntoNotesnbm dataset show that our model outperforms state-of-the-art sentencelevel and document-level NER methods.

PDF Details

AAAI Conference 2015 Conference Paper

Content-Based Collaborative Filtering for News Topic Recommendation

Zhongqi Lu
Zhicheng Dou
Jianxun Lian
Xing Xie
Qiang Yang

News recommendation has become a big attraction with which major Web search portals retain their users. Two effective approaches are Content-based Filtering and Collaborative Filtering, each serving a specific recommendation scenario. The Content-based Filtering approaches inspect rich contexts of the recommended items, while the Collaborative Filtering approaches predict the interests of long-tail users by collaboratively learning from interests of related users. We have observed empirically that, for the problem of news topic displaying, both the rich context of news topics and the long-tail users exist. Therefore, in this paper, we propose a Content-based Collaborative Filtering approach (CCF) to bring both Content-based Filtering and Collaborative Filtering approaches together. We found that combining the two is not an easy task, but the benefits of CCF are impressive. On one hand, CCF makes recommendations based on the rich contexts of the news. On the other hand, CCF collaboratively analyzes the scarce feedbacks from the long-tail users. We tailored this CCF approach for the news topic displaying on the Bing front page and demonstrated great gains in attracting users. In the experiments and analyses part of this paper, we discuss the performance gains and insights in news topic recommendation in Bing.

PDF Details