Papers by Zhicheng Dou

CoRanking: Collaborative Ranking with Small and Large Ranking Agents (2025.findings-emnlp)

Challenge:	Dense retrieval requires high-quality text sequence embeddings to support effective search in the representation space.
Approach:	They propose a self-learning method that pre-trains the autoencoder using a weak decoder to push the encoder to provide better sequence representations.
Outcome:	The proposed model significantly boosts the effectiveness and few-shot ability of dense retrieval models on web search, news recommendation, and open domain question answering.

Copied to clipboard

Wenhan Liu, Xinyu Ma, Yutao Zhu, Lixin Su, Shuaiqiang Wang, Dawei Yin, Zhicheng Dou

CORAL: Benchmarking Multi-turn Conversational Retrieval-Augmented Generation (2025.findings-naacl)

Challenge:	Listwise ranking based on Large Language Models (LLMs) has achieved state-of-the-art performance in Information Retrieval (IR) however, their effectiveness often depends on LLMs with massive parameter scales and computationally expensive sliding window processing, leading to substantial efficiency bottlenecks.
Approach:	They propose a Collaborative Ranking framework (CoRanking) for LLM-based listwise ranking based on large language models with massive parameter scales and computationally expensive sliding window processing.
Outcome:	The proposed framework reduces ranking latency by approximately 70% while improving effectiveness compared to the standalone large reranker.

Copied to clipboard

Yiruo Cheng, Kelong Mao, Ziliang Zhao, Guanting Dong, Hongjin Qian, Yongkang Wu, Tetsuya Sakai, Ji-Rong Wen, Zhicheng Dou

RAG-Studio: Towards In-Domain Adaptation of Retrieval Augmented Generation Through Self-Alignment (2024.findings-emnlp)

Challenge:	Existing research focuses on single-turn RAG, leaving a gap in addressing multi-turn conversations . a new benchmark is designed to assess RAG systems in realistic multi-turned conversations based on Wikipedia .
Approach:	They propose a large-scale benchmark to assess RAG systems in multi-turn contexts . CORAL includes diverse information-seeking conversations automatically derived from Wikipedia . authors propose unified framework to standardize various conversational RAG methods .
Outcome:	The proposed framework supports three core tasks of conversational RAG: passage retrieval, response generation, and citation labeling.

Copied to clipboard

Kelong Mao, Zheng Liu, Hongjin Qian, Fengran Mo, Chenlong Deng, Zhicheng Dou

EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis (2026.findings-acl)

Challenge:	Existing RAG systems that use pre-trained LLMs and retrievers often fail in specialized domains and applications.
Approach:	They propose a self-aligned training framework that adapts general RAG models to specific domains solely through synthetic data.
Outcome:	Experiments on specialized domain corpus, general LLM, and general retriever show that the self-aligned training framework outperforms human-annotated training data in specialized fields.

Copied to clipboard

Xiaoshuai Song, Haofei Chang, Guanting Dong, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou

Defending against Indirect Prompt Injection by Instruction Detection (2025.findings-emnlp)

Challenge:	Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but access to real systems is often restricted and manually built sandboxes are hard to scale.
Approach:	They propose an automated framework for scalable tool-interaction environments via programmatic synthesis that synthesizes 191 environments and about 7K scenarios and applies them to Supervised Fine-Tuning and Reinforcement Learning for Qwen3 series models.
Outcome:	The proposed framework significantly improves LLMs’ ability to solve tasks in complex environments involving multi-turn, multi-tool interactions.

Copied to clipboard

Tongyu Wen, Chenglong Wang, Xiyuan Yang, Haoyu Tang, Yueqi Xie, Lingjuan Lyu, Zhicheng Dou, Fangzhao Wu

An Element is Worth a Thousand Words: Enhancing Legal Case Retrieval by Incorporating Legal Elements (2024.findings-acl)

Challenge:	Indirect Prompt Injection attacks can be exploited by LLMs that are embedded with external data.
Approach:	They propose a detection-based approach that leverages the behavioral states of LLMs to identify potential IPI attacks.
Outcome:	The proposed approach reduces the success rate of attacks to 0.03% on the BIPIA benchmark.

Copied to clipboard

Chenlong Deng, Zhicheng Dou, Yujia Zhou, Peitian Zhang, Kelong Mao

R^3AG: Retriever Routing for Retrieval-Augmented Generation (2026.acl-long)

Challenge:	Existing methods for legal case retrieval lack the definition of relevance for legal cases . however, the definition goes beyond the common semantic relevance of ad-hoc retrieval.
Approach:	They propose a legal element dataset that incorporates legal elements into a semi-automatic method . they propose two models to enhance legal search using legal elements .
Outcome:	The proposed models outperform existing methods in enhancing legal search using legal elements.

Copied to clipboard

Tong Zhao, Yutao Zhu, Yucheng Tian, Zhicheng Dou

Hence, Socrates is mortal: A Benchmark for Natural Language Syllogistic Reasoning (2023.findings-acl)

Challenge:	Retrieval-augmented generation (RAG) is often bottlenecked by the “one-size-fits-all” retrieval paradigm, as different queries exhibit distinct preferences for different retrievers.
Approach:	They propose a novel routing framework that explicitly models the dynamic alignment between queries and retriever capabilities and decomposes retriever capability into two learnable dimensions: retrieval quality and generation utility.
Outcome:	Experiments on knowledge-intensive tasks show that R3AG outperforms both the best individual retrievers and state-of-the-art static routing methods.

Copied to clipboard

Yongkang Wu, Meng Han, Yutao Zhu, Lei Li, Xinyu Zhang, Ruofei Lai, Xiaoguang Li, Yuanhang Ren, Zhicheng Dou, Zhao Cao

FinSight: Towards Real-World Financial Deep Research (2026.acl-long)

Challenge:	SylloBase is a benchmark for syllogistic reasoning, a critical capability widely required in natural language understanding tasks, such as text entailment and question answering.
Approach:	They propose to use a benchmark to learn syllogistic reasoning on a set of templates and to use them to generate and understand slogisms.
Outcome:	The proposed benchmark covers a complete taxonomy of syllogism reasoning patterns, and contains both automatically and manually constructed samples.

Copied to clipboard

Jiajie Jin, Yuyao Zhang, Yimeng Xu, Hongjin Qian, Yutao Zhu, Zhicheng Dou

A Multi-Task Embedder For Retrieval Augmented LLMs (2024.acl-long)

Challenge:	FinSight is the first multi-agent framework for automating end-to-end professional, multimodal financial reports.
Approach:	They propose a code agent with variable memory architecture that unifies data, tools, and agents into a programmable variable space.
Outcome:	The proposed framework outperforms leading deep research systems in factual accuracy, analytical depth, and presentation quality.

Copied to clipboard

Peitian Zhang, Zheng Liu, Shitao Xiao, Zhicheng Dou, Jian-Yun Nie

RetroLLM: Empowering Large Language Models to Retrieve Fine-grained Evidence within Generation (2025.acl-long)

Challenge:	Existing retrieval methods face limitations in terms of knowledge, memory, and action.
Approach:	They propose a retrieval enhancement mechanism that brings in useful information from external sources to augment the LLM.
Outcome:	The proposed method significantly improves the LLM’s performance in various downstream tasks while introducing superior retrieval augmentation’s effect over both general and task-specifc retrievers.

Copied to clipboard

Xiaoxi Li, Jiajie Jin, Yujia Zhou, Yongkang Wu, Zhonghua Li, Ye Qi, Zhicheng Dou

INTERS: Unlocking the Power of Large Language Models in Search with Instruction Tuning (2024.acl-long)

Challenge:	Existing methods rely on separate retrievers to fetch top-k text chunks for generating evidence, and they lack joint optimization.
Approach:	They propose a framework that integrates retrieval and generation into a single, auto-regressive process, enabling LLMs to directly generate fine-grained evidence from the corpus with constrained decoding.
Outcome:	Extensive experiments on five open-domain QA datasets demonstrate the proposed framework’s superior performance across both in-domain and out-of-domain tasks.

Copied to clipboard

Yutao Zhu, Peitian Zhang, Chenghao Zhang, Yifei Chen, Binyu Xie, Zheng Liu, Ji-Rong Wen, Zhicheng Dou

BIDER: Bridging Knowledge Inconsistency for Efficient Retrieval-Augmented LLMs via Key Supporting Evidence (2024.findings-acl)

Challenge:	Large language models (LLMs) have demonstrated impressive capabilities in various natural language processing tasks, but their application to information retrieval tasks is still challenging due to the infrequent occurrence of many IR-specific concepts in natural language.
Approach:	They propose to use instruction tuning to enhance LLMs' proficiency in IR tasks by combining a dataset with manually written templates to analyze the effects of instruction design, template diversity, few-shot demonstrations, and the volume of instructions.
Outcome:	The proposed model can be used to perform query understanding, document understanding, and query-document relationship understanding tasks.

Copied to clipboard

Jiajie Jin, Yutao Zhu, Yujia Zhou, Zhicheng Dou

Green & Sustainable NLP

Challenge:	Large language models (LLMs) have demonstrated efficacy in knowledge-intensive tasks such as open-domain QA, but inconsistencies between retrieval knowledge and the necessary knowledge for LLMs, leading to a decline in LLM’s answer quality.
Approach:	They propose a retrieval-augmented large language model that refines retrieval documents into Key Supporting Evidence (KSE) through knowledge synthesis, supervised fine-tuning, and preference alignment.
Outcome:	The proposed model outperforms existing methods in five datasets and boosts answer quality by 7% while reducing input content length by 80%.

ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability (2026.acl-long)

Copied to clipboard

Wenhan Liu, Xinyu Ma, Weiwei Sun, Yutao Zhu, Yuchen Li, Dawei Yin, Zhicheng Dou

ChatRetriever: Adapting Large Language Models for Generalized and Robust Conversational Dense Retrieval (2024.emnlp-main)

Challenge:	Existing rerankers perform poorly in complex ranking scenarios due to the scarcity of reasoning-intensive training data.
Approach:	They propose an automated reasoning-intensive training framework which generates high-quality training labels from training queries and passages.
Outcome:	The proposed model outperforms baselines significantly and achieves much lower latency than the pointwise reranker.

Copied to clipboard

Kelong Mao, Chenlong Deng, Haonan Chen, Fengran Mo, Zheng Liu, Tetsuya Sakai, Zhicheng Dou

Explicit Query Rewriting for Conversational Dense Retrieval (2022.emnlp-main)

Challenge:	a conversational search system requires accurate interpretation of user intent from complex multi-turn contexts.
Approach:	They propose a dual-learning approach that adapts LLMs for retrieval via contrastive learning while enhancing the complex session understanding through masked instruction tuning.
Outcome:	The proposed approach outperforms existing retrieval methods on five conversational search benchmarks.

Copied to clipboard

Hongjin Qian, Zhicheng Dou

Hierarchical Document Refinement for Long-context Retrieval-augmented Generation (2025.acl-long)

Challenge:	In a conversational search scenario, a query might be context-dependent because some words are referred to previous expressions or omitted.
Approach:	They propose a model that performs query rewriting and context modelling in a unified framework by highlighting relevant terms in the query context.
Outcome:	The proposed model outperforms baseline models in terms of quality of query rewriting and quality of contextualized query embedding.

Copied to clipboard

Jiajie Jin, Xiaoxi Li, Guanting Dong, Yuyao Zhang, Yutao Zhu, Yongkang Wu, Zhonghua Li, Ye Qi, Zhicheng Dou

Enhancing LLM Text Detection with Retrieved Contexts and Logits Distribution Consistency (2025.emnlp-main)

Challenge:	Real-world RAG applications often encounter long-context input scenarios where redundant information and noise results in higher inference costs and reduced performance.
Approach:	They propose an efficient plug-and-play refiner that leverages the structural characteristics of long documents.
Outcome:	Experiments on seven QA datasets show that LongRefiner achieves competitive performance in various scenarios while using 10x fewer computational costs and latency compared to baseline.

Copied to clipboard

Zhaoheng Huang, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou

Search-o1: Agentic Search-Enhanced Large Reasoning Models (2025.emnlp-main)

Challenge:	Existing methods for detecting large language models (LLMs) generate fluent text, but they only use a few tokens due to the short length or insufficient information in some texts.
Approach:	They propose a method that leverages external text corpora to evaluate the difference in logit distribution of input text under retrieved human-written and LLM-rewritten contexts.
Outcome:	The proposed method achieves state-of-the-art performance in AUROC on five public datasets with three widely-used source LLMs.

Copied to clipboard

Xiaoxi Li, Guanting Dong, Jiajie Jin, Yuyao Zhang, Yujia Zhou, Yutao Zhu, Peitian Zhang, Zhicheng Dou

Agentic-R: Learning to Retrieve for Agentic Search (2026.findings-acl)

Challenge:	Large reasoning models (LRMs) have demonstrated impressive long stepwise reasoning capabilities through large-scale reinforcement learning.
Approach:	They propose a framework that enhances large reasoning models with an agentic retrieval-augmented generation mechanism and a Reason-in-Documents module for refining retrieved documents.
Outcome:	The proposed framework enhances LRMs with an agentic retrieval-augmented generation mechanism and Reason-in-Documents module for refining retrieved documents.

Copied to clipboard

Wenhan Liu, Xinyu Ma, Yutao Zhu, Yuchen Li, Daiting Shi, Dawei Yin, Zhicheng Dou

FineRAG: Fine-grained Retrieval-Augmented Text-to-Image Generation (2025.coling-main)

Challenge:	Existing retrievers for single-turn retrieval-augmented generation (RAG) rely on similarity-based retrievers, but similar passages are not always useful for final answer generation.
Approach:	They propose a retrieval-augmented-generation retriever that integrates reasoning with retrieval . they use local query-passage relevance and global answer correctness to measure passage utility .
Outcome:	The proposed retriever outperforms existing retrievers on QA benchmarks on seven single-hop and multi-hop searches.

Copied to clipboard

Huaying Yuan, Ziliang Zhao, Shuting Wang, Shitao Xiao, Minheng Ni, Zheng Liu, Zhicheng Dou

Value Compass Benchmarks: A Comprehensive, Generative and Self-Evolving Platform for LLMs’ Value Evaluation (2025.acl-demo)

Challenge:	Recent advances in text-to-image generation still exhibit limitations in terms of knowledge access.
Approach:	They propose a fine-grained retrieval-augmented image generation model that breaks down the retrieval task into four critical stages: query decomposition, candidate selection, retrieval augmented diffusion, and self-reflection.
Outcome:	The proposed method significantly reduces noise associated with retrieval-augmented image generation and performs better in complex, open-world scenarios.

Copied to clipboard

Jing Yao, Xiaoyuan Yi, Shitong Duan, Jindong Wang, Yuzhuo Bai, Muhua Huang, Yang Ou, Scarlett Li, Peng Zhang, Tun Lu, Zhicheng Dou, Maosong Sun, James Evans, Xing Xie

Challenge:	Current evaluation methods for large language models face two key challenges: 1. evaluation validity and 2. Result interpretation reduce the pluralistic and incommensurable values to one-dimensional scores.
Approach:	They propose a platform for comprehensive value diagnosis of large language models (LLMs) that provides a generative evaluation paradigm that automatically creates real-world test items co-evolving with ever-advancing LLMs.
Outcome:	The proposed platform provides a framework for comprehensive value diagnosis of large language models (LLMs) with fine-grained scores and case studies across 27 value dimensions for 33 leading LLMs, customized comparisons, and visualized analysis of LLM’s alignment with cultural values.

e5-omni: Explicit Cross-modal Alignment for Omni-modal Embeddings (2026.findings-acl)

Copied to clipboard

Haonan Chen, Sicheng Gao, Radu Timofte, Tetsuya Sakai, Zhicheng Dou

ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration (2026.acl-long)

Challenge:	Recent omni-modal embeddings rely heavily on implicit alignment from pretrained visionlanguage models.
Approach:	They propose a lightweight explicit alignment recipe that adapts off-the-shelf VLMs into robust omni-modal embedding models.
Outcome:	The proposed model improves on MMEB-V2 and AudioCaps with a lightweight explicit alignment recipe.

Copied to clipboard

Yifei Chen, Guanting Dong, Zhicheng Dou

Grounding Language Model with Chunking-Free In-Context Retrieval (2024.acl-long)

Challenge:	Existing training frameworks for Large Language Models (LLMs) focus on answers’ accuracy, overlooking specific alignment for behavior patterns.
Approach:	They propose a training framework for calibrating agent’s tool-use behavior through two synergistic perspectives: self-evolving data flywheel and behavior calibration training.
Outcome:	The proposed framework improves the accuracy, efficiency, reasoning conciseness, and tool execution accuracy of large language models.

Copied to clipboard

Hongjin Qian, Zheng Liu, Kelong Mao, Yujia Zhou, Zhicheng Dou

Learning Dynamic Multi-attribute Interest for Personalized Product Search (2024.findings-emnlp)

Challenge:	CFIC retrieval approach eliminates the need for document chunking and provides a more efficient and efficient method for RAG systems.
Approach:	They propose a Chunking-Free In-Context retrieval approach specifically tailored for RAG systems . they employ auto-aggressive decoding to accurately identify specific evidence text .
Outcome:	The proposed method is better than traditional methods on open question answering datasets.

Copied to clipboard

Yutong Bai, Zhicheng Dou, Ji-Rong Wen

Learning Interpretable Legal Case Retrieval via Knowledge-Guided Case Reformulation (2024.emnlp-main)

Challenge:	Existing methods to capture valuable features for Personalized product search ignore that the user’s attention varies on product attributes.
Approach:	They propose a dynamic multi-attribute interest learning model to tackle the influences from attributes to user interests.
Outcome:	The proposed model significantly improves existing methods on large-scale datasets.

Copied to clipboard

Chenlong Deng, Kelong Mao, Zhicheng Dou

Sliding Windows Are Not the End: Exploring Full Ranking with Long-Context Large Language Models (2025.acl-long)

Challenge:	Existing methods for legal case retrieval often overlook the incorporation of legal expert knowledge, leading to unsatisfactory retrieval performance.
Approach:	They propose a legal knowledge-guided case reformulation approach based on large language models for effective and interpretable legal case retrieval.
Outcome:	The proposed model performs better on complex legal case queries than existing methods.

Copied to clipboard

Wenhan Liu, Xinyu Ma, Yutao Zhu, Ziliang Zhao, Shuaiqiang Wang, Dawei Yin, Zhicheng Dou

Web Sitemap Knowledge Can Enhance Autonomous Browsing (2026.findings-acl)

Challenge:	Existing methods for listwise passage ranking use sliding window approach, which is inefficient as it requires repetitive and serialized processing.
Approach:	They propose a listwise label construction approach and importance-aware learning objective for full ranking.
Outcome:	The proposed method outperforms existing methods in listwise ranking tasks.

Copied to clipboard

Yuyao Zhang, Hongyu Lu, Jiajie Jin, Hongjin Qian, Shiyu Li, Zhao Yang, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou

How Credible Is an Answer From Retrieval-Augmented LLMs? Investigation and Evaluation With Multi-Hop QA (2025.coling-main)

Challenge:	Existing web agents suffer from limited robustness, efficiency and task success due to lack of structural understanding of websites and lack of browsing priors in pre-trained models.
Approach:	They propose an agent-oriented sitemap protocol that integrates structured website knowledge into web agents.
Outcome:	The proposed agent-oriented sitemap improves robustness, efficiency and effectiveness without extra training.

Copied to clipboard

Yujia Zhou, Zheng Liu, Zhicheng Dou

RAG-Critic: Leveraging Automated Critic-Guided Agentic Workflow for Retrieval Augmented Generation (2025.acl-long)

Challenge:	Retrieval-augmented large language models (RaLLMs) are reshaping knowledge acquisition, offering long-form, knowledge-grounded answers through advanced reasoning and generation capabilities.
Approach:	They propose a benchmarking system to evaluate RaLLMs' correctness and Groundedness to determine their reliability in multi-hop question-answering tasks.
Outcome:	The proposed model-based evaluation pipeline for multi-hop question-answering tasks reveals that the model generates inaccuracies when dealing with flawed or partial knowledge.

Copied to clipboard

Guanting Dong, Jiajie Jin, Xiaoxi Li, Yutao Zhu, Zhicheng Dou, Ji-Rong Wen

P3: Prompts Promote Prompting (2025.findings-acl)

Challenge:	Recent advances in large language models (LLMs) have demonstrated remarkable performance across a wide range of downstream tasks.
Approach:	They propose a framework that leverages a critic-guided agentic workflow to improve RAG capabilities autonomously.
Outcome:	The proposed framework improves RAG capabilities autonomously by leveraging a critic-guided agentic workflow.

Copied to clipboard

Xinyu Zhang, Yuanquan Hu, Fangchao Liu, Zhicheng Dou

Language Models

Challenge:	Recent advances in prompt optimization have shown effectiveness of using multiple components to optimize models . however, such unilateral approaches often yield suboptimal results due to interdependent nature of these components.
Approach:	They propose a self-improvement framework that optimizes both system and user prompts . they use offline optimized prompts to promote online prompt optimization .
Outcome:	The proposed framework improves performance on general and reasoning tasks.

Small Models, Big Insights: Leveraging Slim Proxy Models To Decide When and What to Retrieve for LLMs (2024.acl-long)

Copied to clipboard

Jiejun Tan, Zhicheng Dou, Yutao Zhu, Peidong Guo, Kun Fang, Ji-Rong Wen

Single LLM, Multiple Roles: A Unified Retrieval-Augmented Generation Framework Using Role-Specific Token Optimization (2025.emnlp-main)

Challenge:	Existing methods to determine the knowledge an LLM already possesses and the knowledge that requires the help of a search engine are expensive and require excessive computational costs.
Approach:	They propose a slim proxy model that detects missing knowledge in LLMs with a proxy model and use it to perform retrieval for the missing knowledge.
Outcome:	The proposed approach detects missing knowledge in LLMs with a slim proxy model and takes its answers as heuristic answers.

Copied to clipboard

Yutao Zhu, Jiajie Jin, Hongjin Qian, Zheng Liu, Zhicheng Dou, Ji-Rong Wen

RLSeek: Evidence-Grounded Reasoning for RAG Hallucination Detection (2026.acl-long)

Challenge:	Existing studies have optimized retrieval-augmented generation (RAG) across sub-tasks, but integrating these optimizations into a unified framework remains challenging.
Approach:	They propose a unified retrieval-augmented generation framework that optimizes role-specific tokens for multi-task processing.
Outcome:	The proposed framework achieves efficient multi-task processing through role-specific token optimization.

Copied to clipboard

Zhaoheng Huang, Dacheng Wen, Yutao Zhu, Xiaoying Lian, Yushi Liang, Kai Hao, Nan Li, Liangjie Zhang, Qi Zhang, Ji-Rong Wen, Zhicheng Dou, Fangzhao Wu

Large Language Models Know Your Contextual Search Intent: A Prompting Framework for Conversational Search (2023.findings-emnlp)

Challenge:	Recent work addresses this problem by training span-level hallucination detectors using reinforcement learning and chain-of-thought reasoning.
Approach:	They propose a framework that explicitly enforces active evidence seeking during CoT reasoning by requiring quotation of relevant source segments at each verification step.
Outcome:	The proposed framework improves hallucination span detection performance with limited reasoning overhead and improved robustness in out-of-domain settings.

Copied to clipboard

Kelong Mao, Zhicheng Dou, Fengran Mo, Jiewen Hou, Haonan Chen, Hongjin Qian

Enhancing Reranking for Recommendation with LLMs through User Preference Retrieval (2025.coling-main)

Challenge:	Existing methods for understanding users’ contextual search intent show unsatisfactory effectiveness and robustness to handle real conversational search scenarios.
Approach:	They propose to use large language models to generate multiple query rewrites and hypothetical responses and to aggregate them into an integrated representation that can robustly represent the user’s real contextual search intent.
Outcome:	The proposed framework can generate multiple query rewrites and hypothetical responses and can be used to represent the user’s real contextual search intent.

Copied to clipboard

Haobo Zhang, Qiannan Zhu, Zhicheng Dou

Language Models

Challenge:	Existing large language models (LLMs) generate redundant output, which generates irrelevant information about the user’s preferences on candidate items from user behavior sequences.
Approach:	They propose a framework that enhances reranking for recommendation with large language models through user preference retrieval.
Outcome:	The proposed framework improves reranking for recommendation with large language models through user preference retrieval on three real-world public datasets.

Towards Effective and Efficient Continual Pre-training of Large Language Models (2025.acl-long)

Copied to clipboard

Jie Chen, Zhipeng Chen, Jiapeng Wang, Kun Zhou, Yutao Zhu, Jinhao Jiang, Yingqian Min, Xin Zhao, Zhicheng Dou, Jiaxin Mao, Yankai Lin, Ruihua Song, Jun Xu, Xu Chen, Rui Yan, Zhewei Wei, Di Hu, Wenbing Huang, Ji-Rong Wen

Language Models

Challenge:	Continual pre-training (CPT) has been an important approach for adapting language models to specific domains or tasks.
Approach:	They propose a Continual pre-training method that can greatly improve Chinese language ability and scientific reasoning ability of LLMs.
Outcome:	The proposed method can greatly improve Chinese language ability and scientific reasoning ability of LLMs.

Coarse-to-Fine: Hierarchical Multi-task Learning for Natural Language Understanding (2022.coling-1)

Copied to clipboard

Zhaoye Fei, Yu Tian, Yongkang Wu, Xinyu Zhang, Yutao Zhu, Zheng Liu, Jiawen Wu, Dejiang Kong, Ruofei Lai, Zhao Cao, Zhicheng Dou, Xipeng Qiu

Less is More: Learning to Refine Dialogue History for Personalized Dialogue Generation (2022.naacl-main)

Challenge:	Existing methods to learn downstream tasks by stitches skill block lack rationality and interpretation.
Approach:	They propose a hierarchical framework with a coarse-to-fine paradigm for generalized text representations from the large-scale corpus.
Outcome:	The proposed model learns basic language properties from all tasks and boosts performance on relevant tasks.

Copied to clipboard

Hanxun Zhong, Zhicheng Dou, Yutao Zhu, Hongjin Qian, Ji-Rong Wen

Dialogue Systems & Conversational Agents

Challenge:	Existing personalized dialogue systems extract user profiles from dialogue history to guide personalized response generation.
Approach:	They propose to refine the user dialogue history on a large scale to obtain more persona information from the dialogue history and leverage other similar users' data to enhance personalization.
Outcome:	The proposed model can handle more dialogue history and obtain more abundant and accurate persona information.

OmniEval: An Omnidirectional and Automatic RAG Evaluation Benchmark in Financial Domain (2025.emnlp-main)

Copied to clipboard

Shuting Wang, Jiejun Tan, Zhicheng Dou, Ji-Rong Wen

ToolScope: An Agentic Framework for Vision-Guided and Long-Horizon Tool Use (2026.findings-acl)

Challenge:	a new benchmark for RAG is developed for the financial domain . omnidirectional and automatic benchmarks are difficult to build in vertical domains .
Approach:	They propose an omnidirectional and automatic RAG benchmark for the financial domain . they categorize RAG scenarios by task classes and 16 financial topics .
Outcome:	The proposed benchmark achieves an 87.47% acceptance ratio in human evaluations of generated instances.

Copied to clipboard

Mengjie Deng, Guanting Dong, Zhicheng Dou

LLM-Generated Text May Harm Your Retrieval! A Robust Detection Strategy for Retrieval-Augmented Generation (2026.acl-long)

Challenge:	Recent advances in large language models have demonstrated remarkable problem-solving capabilities . however, enabling multimodal large language model to flexibly and efficiently utilize external tools remains a challenge .
Approach:	They introduce an agentic framework to unify global planning with local multimodal perception . they evaluate ToolScope on four VQA benchmarks across diverse domains .
Outcome:	The proposed framework unifies global planning with local multimodal perception . it adopts a specialized Perceive tool to mitigate visual context degradation in long-horizon VQA task.

Copied to clipboard

Zhaoheng Huang, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou

Enabling Discriminative Reasoning in LLMs for Legal Judgment Prediction (2024.findings-emnlp)

Challenge:	Retrieval-augmented generation (RAG) improves accuracy and timeliness of large language models, but external corpora may become contaminated with LLM-generated texts.
Approach:	They propose a method that integrates external knowledge retrieved from external sources into RAG to filter out LLM-generated texts from retrieved results.
Outcome:	The proposed method mitigates performance degradation and improves stability of RAG systems.

Copied to clipboard

Chenlong Deng, Kelong Mao, Yuyao Zhang, Zhicheng Dou

ConvTrans: Transforming Web Search Sessions for Conversational Dense Retrieval (2022.emnlp-main)

Challenge:	Existing large language models (LLMs) underperform in legal judgment prediction due to challenges in understanding case facts and distinguishing between similar charges.
Approach:	They propose a framework that allows LLMs to discriminate among charges and a judicial reasoning framework to improve their models for effective legal judgment prediction.
Outcome:	The proposed framework improves accuracy and efficiency when dealing with complex and confusing charges.

Copied to clipboard

Kelong Mao, Zhicheng Dou, Hongjin Qian, Fengran Mo, Xiaohua Cheng, Zhao Cao

Boosting Long-Context Information Seeking via Query-Guided Activation Refilling (2025.acl-long)

Challenge:	Recent studies show that conversational dense retrieval is a promising technique for realizing conversational search, but its implementation is severely hindered by the lack of data.
Approach:	They propose a method that transforms easily-accessible web search sessions into conversational search sessions to alleviate the data scarcity problem.
Outcome:	The proposed method can transform easily-accessible web search sessions into conversational search sessions.

Copied to clipboard

Hongjin Qian, Zheng Liu, Peitian Zhang, Zhicheng Dou, Defu Lian

Green & Sustainable NLP

Challenge:	Existing methods for processing long contexts are ineffective due to their inherent context window limitations and the computational burden of extensive key-value activations.
Approach:	They propose a method for processing long context information-seeking tasks via query-guided ACtivation REfilling (ACRE) a bi-layer KV Cache is constructed where the layer-1 cache compactly captures global information and the layer-2 cache provides detailed, localized information.
Outcome:	The proposed method achieves significant improvements in both performance and efficiency on a variety of long-context information-seeking datasets.

MCP: Self-supervised Pre-training for Personalized Chatbots with Multi-level Contrastive Sampling (2022.findings-emnlp)

Copied to clipboard

Zhaoheng Huang, Zhicheng Dou, Yutao Zhu, Zhengyi Ma

Dialogue Systems & Conversational Agents

Challenge:	Existing studies focus on generating implicit user profiles from the user’s dialogue history, thus it suffers from data sparsity and performance degradation.
Approach:	They propose a self-supervised learning framework MCP for capturing better representations from users’ dialogue history for personalized chatbots.
Outcome:	The proposed model improves on two real-world datasets.

Interpreting Conversational Dense Retrieval by Rewriting-Enhanced Inversion of Session Embedding (2024.acl-long)

Copied to clipboard

Yiruo Cheng, Kelong Mao, Zhicheng Dou

Explainability & Interpretability in NLP

Challenge:	Conversational dense retrieval models lack interpretability, hindering intuitive understanding of model behaviors . a major limitation of conversational dense search is their lack of interpretability .
Approach:	They propose to transform opaque session embeddings into explicit interpretable text . they propose to incorporate external interpretable query rewrites into the transformation process .
Outcome:	The proposed approach yields more interpretable text and preserves original retrieval performance over baselines.

Little Giants: Synthesizing High-Quality Embedding Data at Scale (2025.naacl-long)

Copied to clipboard

Haonan Chen, Liang Wang, Nan Yang, Yutao Zhu, Ziliang Zhao, Furu Wei, Zhicheng Dou

Generalizing Conversational Dense Retrieval via LLM-Cognition Data Augmentation (2024.acl-long)

Challenge:	Synthetic data generation is an increasingly popular way of training models without the need for large, manually labeled datasets.
Approach:	They propose a framework that aligns open-source small models to efficiently generate large-scale embedding data.
Outcome:	The proposed framework outperforms state-of-the-art embedding models by using only 1/10 of the GPT API calls.

Copied to clipboard

Haonan Chen, Zhicheng Dou, Kelong Mao, Jiongnan Liu, Ziliang Zhao

Hybrid Inverted Index Is a Robust Accelerator for Dense Retrieval (2023.emnlp-main)

Challenge:	Existing conversational dense retrieval models view a conversation as a fixed sequence of questions and responses, and these alternate conversations are unrecorded.
Approach:	They propose a framework for generalizing Conversational dense retrieval via LLM-cognition data Augmentation (ConvAug) they first generate multi-level augmented conversations to capture the diverse nature of conversational contexts.
Outcome:	The proposed framework generalizes Conversational dense retrieval via LLM-cognition data Augmentation on four public datasets.

Copied to clipboard

Peitian Zhang, Zheng Liu, Shitao Xiao, Zhicheng Dou, Jing Yao

Neuro-Symbolic Query Compiler (2025.findings-acl)

Challenge:	Inverted file structure is a common technique for accelerating dense retrieval, but its lossy nature degrades it.
Approach:	They propose a hybrid index where embedding clusters and salient terms work collaboratively to accelerate dense retrieval.
Outcome:	The proposed method achieves lossless retrieval quality with competitive efficiency across index settings.

Copied to clipboard

Yuyao Zhang, Zhicheng Dou, Xiaoxi Li, Jiajie Jin, Yongkang Wu, Zhonghua Li, Ye Qi, Ji-Rong Wen

Search-Oriented Conversational Query Editing (2023.findings-acl)

Challenge:	Retrieval-Augmented Generation (RAG) systems are limited in their ability to process information in open-source environments.
Approach:	They propose a neuro-symbolic framework inspired by linguistic grammar rules and compiler design to formalize complex queries using a minimal yet sufficient Backus-Naur Form grammar.
Outcome:	The proposed framework is based on a backus-naur form grammar and compiler design that maintains completeness while minimizing redundancy.

Copied to clipboard

Kelong Mao, Zhicheng Dou, Bang Liu, Hongjin Qian, Fengran Mo, Xiangli Wu, Xiaohua Cheng, Zhao Cao

GLARE: Agentic Reasoning for Legal Judgment Prediction (2026.acl-long)

Challenge:	Existing CQR models are not learned toward improving the downstream search performance . existing models generate the rewrite token-by-token from scratch .
Approach:	They propose a text editing-based CQR model tailored for conversational search . they propose rewrite tokens are selected from the dialogue in a non-autoregressive fashion .
Outcome:	The proposed model outperforms state-of-the-art models on three conversational search benchmarks while having low rewriting latency.

Copied to clipboard

Xinyu Yang, Chenlong Deng, Zhicheng Dou

Reasoning-Aware AIGC Detection via Alignment and Reinforcement (2026.findings-acl)

Challenge:	Large language models struggle with fine-grained distinctions between similar charges.
Approach:	They propose an agentic legal reasoning framework that actively retrieves external knowledge during decision-making.
Outcome:	The proposed model outperforms baseline models on complex cases involving confusing or rare charges on real-world datasets.

Copied to clipboard

Zhao Wang, Max Xiong, Jianxun Lian, Zhicheng Dou

Progressive Multimodal Reasoning via Active Retrieval (2025.acl-long)

Challenge:	Existing approaches to AIGC detection have relied on statistical classifiers or black-box neural models, which exploit surface-level patterns and struggle to generalize as LLMs evolve.
Approach:	They propose a framework that generates interpretable reasoning chains before classification using supervised fine-tuning and reinforcement learning to improve accuracy.
Outcome:	The proposed framework achieves state-of-the-art performance across multiple benchmarks, offering a robust and transparent solution for AIGC detection.

Copied to clipboard

Guanting Dong, Chenghao Zhang, Mengjie Deng, Yutao Zhu, Zhicheng Dou, Ji-Rong Wen

Memory Matters More: Event-Centric Memory as a Logic Map for Agent Searching and Reasoning (2026.findings-acl)

Challenge:	Existing approaches to improve multimodal large language models' reasoning performance are limited.
Approach:	They propose a framework to progressively improve multimodal reasoning capabilities . they propose active retrieval and Monte Carlo tree search to improve MLLMs' reasoning .
Outcome:	The proposed framework improves multimodal reasoning capabilities in multimodal large language models.

Copied to clipboard

Yuyang Hu, Jiongnan Liu, Jiejun Tan, Yutao Zhu, Zhicheng Dou

A Silver Bullet or a Compromise for Full Attention? A Comprehensive Study of Gist Token-based Context Compression (2025.acl-long)

Challenge:	Existing methods for storing and retrieving memory are limited by shallow semantic retrieval.
Approach:	They propose a memory mechanism that organizes and retrieves past experiences to support decision-making.
Outcome:	Experiments on LoCoMo and NarrativeQA show that CompassMem improves retrieval and reasoning performance across multiple backbone models.

Copied to clipboard

Chenlong Deng, Zhisong Zhang, Kelong Mao, Shuaiyi Li, Xinting Huang, Dong Yu, Zhicheng Dou

Challenge:	gist-based context compression methods can achieve only slight performance loss on tasks like retrieval-augmented generation and long-document QA, but it faces challenges in tasks like synthetic recall.
Approach:	They propose two strategies to improve gist-based context compression in large language models.
Outcome:	The proposed methods can achieve only slight performance loss on retrieval-augmented generation and long-document QA tasks, but they face challenges in tasks like synthetic recall.

RichRAG: Crafting Rich Responses for Multi-faceted Queries in Retrieval-Augmented Generation (2025.coling-main)

Copied to clipboard

Shuting Wang, Xin Yu, Mang Wang, Weipeng Chen, Yutao Zhu, Zhicheng Dou

ATIR: Towards Audio-Text Interleaved Contextual Retrieval (2026.acl-long)

Challenge:	Existing studies focus on question scenarios with clear user intents and concise answers, but it is prevalent that users issue broad, open-ended queries with diverse sub-intents.
Approach:	They propose a framework that includes a sub-aspect explorer and a multi-faceted retriever to build a candidate pool of diverse external documents related to these sub-intents.
Outcome:	The proposed framework provides comprehensive and satisfying responses to users on two publicly available datasets.

Copied to clipboard

Tong Zhao, Chenghao Zhang, Yutao Zhu, Zhicheng Dou

LLMs + Persona-Plug = Personalized LLMs (2025.acl-long)

Challenge:	Recent multimodal information retrieval research has focused on images, largely overlooking audio.
Approach:	They propose an audio-text interleaved contextual retrieval task where queries can alternate between audio and text modalities.
Outcome:	The proposed model significantly improves over baselines.

Copied to clipboard

Jiongnan Liu, Yutao Zhu, Shuting Wang, Xiaochi Wei, Erxue Min, Yu Lu, Shuaiqiang Wang, Dawei Yin, Zhicheng Dou

mmE5: Improving Multimodal Multilingual Embeddings via High-quality Synthetic Data (2025.findings-acl)

Challenge:	Large language models (LLMs) have demonstrated extraordinary capabilities in natural language understanding, generation, and reasoning.
Approach:	They propose a plug-and-play LLM model that embeds a user-specific embedding for each individual by modeling her historical contexts through a lightweight plug-in user embedder module.
Outcome:	Experiments on various tasks in the language model personalization (LaMP) benchmark show that the proposed model significantly outperforms existing personalized LLM approaches.

Copied to clipboard

Haonan Chen, Liang Wang, Nan Yang, Yutao Zhu, Ziliang Zhao, Furu Wei, Zhicheng Dou

MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings (2026.acl-long)

Challenge:	Multimodal embedding models encode multimedia inputs into latent vector representations.
Approach:	They propose to synthesize multimodal multilingual data using a multimodal large language model . they identify three criteria for high-quality synthetic multimodal data .
Outcome:	The proposed model outperforms existing models on the MMEB Benchmark and the XTD benchmark.

Copied to clipboard

Haonan Chen, Hong Liu, Yuping Luo, Liang Wang, Nan Yang, Furu Wei, Zhicheng Dou

ScriptWriter: Narrative-Guided Script Generation (2020.acl-main)

Challenge:	Existing approaches to embed multimodal models face limitations such as suboptimal causal attention in VLMs and limited diversity in training objectives and data.
Approach:	They propose a framework for transforming pre-trained VLMs into bidirectional multimodal embedding models.
Outcome:	The proposed model improves performance across MMEB and ViDoRe-v2 benchmarks and exhibits strong scalability with both model size and training data on MMEF.

Copied to clipboard

Yutao Zhu, Ruihua Song, Zhicheng Dou, Jian-Yun Nie, Jin Zhou

PaperScope: A Multi-Modal Multi-Document Benchmark for Agentic Deep Research Across Massive Scientific Papers (2026.findings-acl)

Challenge:	Existing systems that generate scripts from a storyline are not available for this purpose.
Approach:	They propose a model that generates a story from a narrative and a tool that keeps track of what is said and what is to be said.
Outcome:	The proposed model outperforms baselines that use the narrative as a kind of context.

Copied to clipboard

Lei Xiong, Huaying Yuan, Zheng Liu, Zhao Cao, Zhicheng Dou