Papers by Zhicheng Wang

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

Challenge:	Large Language Models (LLMs) have achieved remarkable performance across NLP tasks . however, in long-context scenarios, they face high computational cost and information redundancy.
Approach:	They propose an encoder-decoder context compression framework that generates a compact sequence of soft tokens for downstream tasks.
Outcome:	Experiments show that GMSA outperforms baselines on multiple long-context question answering and summarization benchmarks while maintaining low end-to-end latency.

Copied to clipboard

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

Responsible & Trustworthy NLP

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.

Rehearsal-free Continual Language Learning via Efficient Parameter Isolation (2023.acl-long)

Copied to clipboard

Zhicheng Wang, Yufang Liu, Tao Ji, Xiaoling Wang, Yuanbin Wu, Congcong Jiang, Ye Chao, Zhencong Han, Ling Wang, Xu Shao, Wenqiu Zeng

Green & Sustainable NLP

Challenge:	Existing methods for learning continual tasks do not cache history data, which makes the problem more challenging.
Approach:	They propose a method that allocates a small portion of private parameters and learns them with a shared pre-trained model.
Outcome:	The proposed method is comparable to existing methods and comparable to those using historical data.

Understanding GUI Agent Localization Biases through Logit Sharpness (2025.findings-emnlp)

Copied to clipboard

Xingjian Tao, Yiwei Wang, Yujun Cai, Zhicheng Yang, Jing Tang

Challenge:	Multimodal large language models often exhibit hallucinations that compromise reliability . despite promising performance, these models often display systematic localization errors .
Approach:	They propose a framework that categorizes model predictions into four distinct types . they propose metric that evaluates alignment between semantic continuity and logits distribution .
Outcome:	The proposed framework categorizes model predictions into four different types . it reveals nuanced failure modes beyond traditional accuracy metrics .

When Inverse Data Outperforms: Exploring the Pitfalls of Mixed Data in Multi-Stage Fine-Tuning (2025.findings-emnlp)

Copied to clipboard

Mengyi Deng, Xin Li, Tingyu Zhu, Zhicheng Yang, Zhijiang Guo, Wei Wang

InquireMobile: Teaching VLM-based Mobile Agent to Request Human Assistance via Reinforcement Fine-Tuning (2026.acl-long)

Challenge:	Existing methods for o1-level performance focus on unidirectional supervised fine-tuning (SFT), overlooking the intricate interplay between diverse reasoning patterns.
Approach:	They construct a reverse reasoning dataset and examine how it is supervised . they find that naively mixing forward and reverse data during SFT weakens the directional distinction .
Outcome:	The proposed model improves accuracy by 1.6%–6.8% over a standard model.

Copied to clipboard

Qihang Ai, Pi Bu, Yue Cao, Yingyao Wang, Jihao Gu, Jingxuan Xing, Zekun Zhu, Wei Jiang, Zhicheng Zheng, Jun Song, Yuning Jiang

Mixture of Heterogeneous Grouped Experts for Language Modeling (2026.acl-industry)

Challenge:	Recent advances in Vision-Language Models (VLMs) have enabled mobile agents to perceive and interact with real-world mobile environments based on human instructions.
Approach:	They propose a vision-language model that actively seeks human confirmation at critical decision points and a model inspired by reinforcement learning.
Outcome:	The proposed model achieves an improvement of 46.8% in inquiry success rate and the best overall success rate among existing baselines on InquireBench.

Copied to clipboard

Zhicheng Ma, Xiang Liu, Zhaoxiang Liu, Ning Wang, Yi Shen, Kai Wang, Shuming Shi, Shiguo Lian

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

Challenge:	Large Language Models (LLMs) based on Mixture-of-Experts (MoE) enforce uniform expert sizes, creating a rigidity that fails to align computational costs with varying token-level complexity.
Approach:	They propose a mixture of heterogeneous grouped experts (MoHGE) that allows for flexible, resource-aware expert combinations.
Outcome:	The proposed model matches the performance of existing Mixture-of-Experts architectures while maintaining balanced GPU utilization.

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.

Prompt-Guided Retrieval Augmentation for Non-Knowledge-Intensive Tasks (2023.findings-acl)

Copied to clipboard

Zhicheng Guo, Sijie Cheng, Yile Wang, Peng Li, Yang Liu

Low-Resource NLP

Challenge:	Recent studies focus on retrieval to solve knowledge-intensive tasks, but the potential of retrieval for non-knowledge-intensive (NKI) tasks remains under-explored.
Approach:	They propose a task-agnostic retrieval framework for NKI tasks that uses a static index and a prompt-guided reranker to re-rank the nearest evidence according to task-specific relevance.
Outcome:	The proposed framework outperforms state-of-the-art retrieval-augmented methods on NKI tasks and will be released for further research.

AlignedCoT: Prompting Large Language Models via Native-Speaking Demonstrations (2024.findings-emnlp)

Copied to clipboard

Zhicheng Yang, Yinya Huang, Jing Xiong, Liang Feng, Xiaodan Liang, Yiwei Wang, Jing Tang

Reasoning

Challenge:	Existing LLMs are delicate and elusive in prompt words and styles.
Approach:	They propose an LLM-acquainted prompting technique that includes proficient "native-speaking" they propose to use in-context learning to prompt LLMs to perform high-performance reasoning .
Outcome:	The proposed technique achieves step-wise prompts in zero-shot scenarios while maintaining the prompt quality.

StableToolBench-MirrorAPI: Modeling Tool Environments as Mirrors of 7,000+ Real-World APIs (2025.findings-acl)

Copied to clipboard

Zhicheng Guo, Sijie Cheng, Yuchen Niu, Hao Wang, Sicheng Zhou, Wenbing Huang, Yang Liu

CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databases (2025.naacl-long)

Challenge:	Existing tool environments face challenges in balancing stability, scale, and realism, especially for benchmarking purposes.
Approach:	They propose a framework that trains specialized LLMs to accurately simulate real API responses by supervised fine-tuning and chain-of-thought reasoning.
Outcome:	The proposed framework achieves superior accuracy and stability compared to state-of-the-art methods on the newly constructed MirrorAPI-Bench and its integration into StableToolBench.

Copied to clipboard

Xiangyan Liu, Bo Lan, Zhiyuan Hu, Yang Liu, Zhicheng Zhang, Fei Wang, Michael Qizhe Shieh, Wenmeng Zhou

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

Challenge:	Large Language Models excel in stand-alone code tasks but struggle with handling entire code repositories.
Approach:	They propose a system that integrates LLM agents with graph database interfaces extracted from code repositories.
Outcome:	The proposed system integrates LLM agents with graph database interfaces extracted from code repositories.

Copied to clipboard

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

Decoding in Latent Spaces for Efficient Inference in LLM-based Recommendation (2025.findings-emnlp)

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

Chengbing Wang, Yang Zhang, Zhicheng Wang, Tianhao Shi, Keqin Bao, Fuli Feng, Tat-Seng Chua

Green & Sustainable NLP

Challenge:	Light Latent-space Decoding (L2D) is an efficient and efficient latent- space decoding method.
Approach:	They propose to bypass language-space decoding by matching candidate items with LLM's internal thought representations in the latent space.
Outcome:	The proposed method is 10x faster than language-space decoding while maintaining or enhancing performance.

Cross-MoE: An Efficient Temporal Prediction Framework Integrating Textual Modality (2025.emnlp-main)

Copied to clipboard

Ruizheng Huang, Zhicheng Zhang, Yong Wang

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

Challenge:	Existing models ignore dynamic and different relations between time series patterns and textual features, which leads to poor performance in temporal-textual feature fusion.
Approach:	They propose a temporal-textual fusion framework that replaces Cross Attention with Cross-Ranker to reduce computational complexity and enhances modality-aware correlation memorization with Mixture-of-Experts (MoE) networks to tolerate the distributional shifts in time series.
Outcome:	The proposed framework reduces MSE by 8.78% compared to the current SOTA model and requires only 75% of computational overhead and 12.5% of activated parameters.

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

StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models (2024.findings-acl)

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.

Copied to clipboard

Zhicheng Guo, Sijie Cheng, Hao Wang, Shihao Liang, Yujia Qin, Peng Li, Zhiyuan Liu, Maosong Sun, Yang Liu

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

Challenge:	Large Language Models (LLMs) have witnessed remarkable advancements in recent years, prompting the exploration of tool learning.
Approach:	They propose a virtual API server and stable evaluation system to assess the stability of large-scale real-time APIs.
Outcome:	The proposed benchmarks demonstrate the stability of the proposed system and its caching system.

Copied to clipboard

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

Information Retrieval

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.

AnchorMem: Anchored Facts with Associative Contexts for Building Memory in Large Language Models (2026.findings-acl)

Copied to clipboard

Zhanyu Shen, Sijie Cheng, Zhicheng Guo, Weiqin Wang, Yile Wang, Hui Huang

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

Challenge:	Existing memory systems rely on summarization to preserve contextual nuances and obscuring key retrieval features.
Approach:	They propose a method that decouples the retrieval unit from the generation context.
Outcome:	The proposed method outperforms baseline models on the LoCoMo benchmark.

Copied to clipboard

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

Protein Large Language Models: A Comprehensive Survey (2025.findings-emnlp)

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

Yijia Xiao, Wanjia Zhao, Junkai Zhang, Yiqiao Jin, Han Zhang, Zhicheng Ren, Renliang Sun, Haixin Wang, Guancheng Wan, Pan Lu, Xiao Luo, Yu Zhang, James Zou, Yizhou Sun, Wei Wang

Lightweight LLM Agent Memory with Small Language Models (2026.acl-long)

Challenge:	Existing studies focus on specific aspects or applications, but this study provides a comprehensive overview of Protein-specific large language models.
Approach:	This paper proposes a structured taxonomy of state-of-the-art ProteinLLMs . they analyze how they leverage large-scale protein sequence data for improved accuracy .
Outcome:	The proposed model covers their architectures, training datasets, evaluation metrics, and diverse applications.

Copied to clipboard

Jiaquan Zhang, Chaoning Zhang, Shuxu Chen, Zhenzhen Huang, Pengcheng Zheng, Zhicheng Wang, Ping Guo, Fan Mo, Sung-Ho Bae, Jie Zou, Jiwei Wei, Yang Yang

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

Challenge:	Existing external memory systems for LLMs have low online overhead but are unstable in accumulating latency over long interactions.
Approach:	They propose a lightweight memory system for better agent memory driven by Small Language Models . lightmem modularizes memory retrieval, writing, and long-term consolidation . they show consistent gains across model scales and high efficiency .
Outcome:	The proposed system improves agent memory but has low latency and low online overhead . it separates online processing from offline consolidation to enable efficient memory invocation . the proposed system achieves an average F1 improvement of 2.5 over A-MEM on LoCoMo .

Copied to clipboard

Zhao Wang, Max Xiong, Jianxun Lian, Zhicheng Dou

Reasoning

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.

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

Information Retrieval

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.

Chinese SimpleQA: A Chinese Factuality Evaluation for Large Language Models (2025.acl-long)

Copied to clipboard

Yancheng He, Shilong Li, Jiaheng Liu, Yingshui Tan, Weixun Wang, Hui Huang, Xingyuan Bu, Hangyu Guo, Chengwei Hu, Boren Zheng, Zhuoran Lin, Dekai Sun, Zhicheng Zheng, Wenbo Su, Bo Zheng

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

Challenge:	Current frontier models sometimes generate false outputs or answers that are not substantiated by evidence.
Approach:	They propose Chinese SimpleQA, a Chinese benchmark to evaluate LLMs' factuality . they focus on Chinese language over 6 major topics with 99 diverse subtopics .
Outcome:	The Chinese SimpleQA benchmark evaluates the factuality ability of LLMs . the questions and answers are short and easy-to-evaluate .

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

VC4VG: Optimizing Video Captions for Text-to-Video Generation (2025.emnlp-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

Yang Du, Zhuoran Lin, Kaiqiang Song, Biao Wang, Zhicheng Zheng, Tiezheng Ge, Bo Zheng, Qin Jin

Improving Speech Translation by Fusing Speech and Text (2023.findings-emnlp)

Challenge:	Recent advances in text-to-video generation highlight the critical role of high-quality video-text pairs in training models capable of producing coherent and instruction-aligned videos.
Approach:	They propose a caption optimization framework tailored to the needs of T2V models.
Outcome:	The proposed framework improves video caption quality and video generation performance.

Copied to clipboard

Wenbiao Yin, Zhicheng Liu, Chengqi Zhao, Tao Wang, Jian Tong, Rong Ye

Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning? (2025.acl-long)

Challenge:	In speech translation, multimodal data to address limitations of individual modalities has shown significant effectiveness.
Approach:	They propose a cross-modal model which supports three input modalities for speech, text and fused speech-text.
Outcome:	The proposed model achieves an average of 34.0 BLEU on MuST-C, GigaST and newstest benchmark.

Copied to clipboard

Yancheng He, Shilong Li, Jiaheng Liu, Weixun Wang, Xingyuan Bu, Ge Zhang, Z.y. Peng, Zhaoxiang Zhang, Zhicheng Zheng, Wenbo Su, Bo Zheng