Papers by Ji Guo

AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments (2026.acl-long)

Challenge:	Existing knowledge graph embedding methods fail to model non-commutative composition patterns . Existing methods are limited to complex space, resulting in a large number of parameters.
Approach:	They propose a knowledge graph embedding method that transforms the coordinates of each entity and then represents each relation as a rotation from head entity to tail entity in complex space.
Outcome:	The proposed method outperforms state-of-the-art methods on link prediction and path query answering.

Copied to clipboard

Zhiheng Xi, Dingwen Yang, Jiaqi Liu, Jixuan Huang, Honglin Guo, Baodai Huang, Tinggang Chen, Qi Zhang, Zhonghang Lu, Chenyu Liu, Jiajun Sun, Jiazheng Zhang, Dingwei Zhu, Xin Guo, Junzhe Wang, Zhihao Zhang, Yuming Yang, Junjie Ye, Minghe Gao, Dongrui Liu, Jiaming Ji, Guohao Li, Tao Gui, Qi Zhang, Xuanjing Huang

Weakly-Supervised Aspect-Based Sentiment Analysis via Joint Aspect-Sentiment Topic Embedding (2020.emnlp-main)

Challenge:	Existing benchmarks evaluate agents in simplified, idealized settings, relying on pre-packaged tool interfaces, overlooking critical steps, and assume inputs are clean and fully specified.
Approach:	They propose a framework that evaluates language agents in simplified, idealized settings . they show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .
Outcome:	Experiments on 15 proprietary and open-source models show that even SOTA systems like Gemini and GPT-5 struggle on AgentGym2 .

Copied to clipboard

Jiaxin Huang, Yu Meng, Fang Guo, Heng Ji, Jiawei Han

Multi-Modal Knowledge Graph Transformer Framework for Multi-Modal Entity Alignment (2023.findings-emnlp)

Challenge:	Existing methods for aspect-based sentiment analysis of review text use only a few keywords describing each aspect/sentiment without using any labeled examples.
Approach:	They propose a weakly-supervised approach for aspect-based sentiment analysis which uses only a few keywords describing each aspect/sentiment without using any labeled examples.
Outcome:	The proposed method generates quality joint topics and outperforms baselines significantly on benchmark datasets.

Copied to clipboard

Qian Li, Cheng Ji, Shu Guo, Zhaoji Liang, Lihong Wang, Jianxin Li

REPT: Bridging Language Models and Machine Reading Comprehension via Retrieval-Based Pre-training (2021.findings-acl)

Challenge:	Multi-modal entity alignment (MMEA) is a critical task that aims to identify equivalent entity pairs across multi-modal knowledge graphs (MMKGs).
Approach:	They propose a novel MMEA transformer that hierarchically introduces neighbor features, multi-modal attributes, and entity types to enhance alignment task.
Outcome:	The proposed transformer hierarchically introduces neighbor features, multi-modal attributes, and entity types to enhance the alignment task.

Copied to clipboard

Fangkai Jiao, Yangyang Guo, Yilin Niu, Feng Ji, Feng-Lin Li, Liqiang Nie

Reasoning

Challenge:	Pre-trained language models have achieved great success on Machine Reading Comprehension (MRC) however, the poor support in evidence extraction hinders them from further advancing MRC.
Approach:	They propose a REtrieval-based pre-training approach that strengthens evidence extraction during pre-training by inherited downstream MRC tasks.
Outcome:	The proposed approach strengthens evidence extraction during pre-training, which is further inherited by downstream tasks.

Hierarchical Context-aware Network for Dense Video Event Captioning (2021.acl-long)

Copied to clipboard

Lei Ji, Xianglin Guo, Haoyang Huang, Xilin Chen

Dual-Gated Fusion with Prefix-Tuning for Multi-Modal Relation Extraction (2023.findings-acl)

Challenge:	Existing technologies for dense video event captioning generate fine-grained captions for all events in a long untrimmed video.
Approach:	They propose a hierarchical context-aware network for dense video event captioning to capture context from various aspects.
Outcome:	The proposed model outperforms the existing model on youcook2 and activitynet . it generates coherent captions for events in a long untrimmed video .

Copied to clipboard

Qian Li, Shu Guo, Cheng Ji, Xutan Peng, Shiyao Cui, Jianxin Li, Lihong Wang

SafeMT: Multi-turn Safety for Multimodal Language Models (2026.acl-long)

Challenge:	Existing methods for multi-modal relation extraction lack useful visual information.
Approach:	They propose a novel multi-modal relation extraction framework to capture deeper correlations of text, entity pair, and image/objects.
Outcome:	The proposed framework captures the deeper correlations of text, entity pair, and image/objects, and extracts useful information.

Copied to clipboard

Han Zhu, Juntao Dai, Jiaming Ji, Haoran Li, Chengkun Cai, Pengcheng Wen, Chi-Min Chan, Boyuan Chen, Yaodong Yang, Sirui Han, Yike Guo

Responsible & Trustworthy NLP

Challenge:	Multi-turn dialogues pose a greater risk than single prompts, but existing safety benchmarks do not account for this situation.
Approach:	They propose a benchmark that features dialogues of varying lengths generated from harmful queries accompanied by images.
Outcome:	The proposed model reduces multi-turn Attack Success Rate (ASR) compared to existing guard models.

Boosting Policy and Process Reward Models with Monte Carlo Tree Search in Open-Domain QA (2025.findings-acl)

Copied to clipboard

Chi-Min Chan, Chunpu Xu, Junqi Zhu, Jiaming Ji, Donghai Hong, Pengcheng Wen, Chunyang Jiang, Zhen Ye, Yaodong Yang, Wei Xue, Sirui Han, Yike Guo

Question Answering

Challenge:	Experimental results show that our approach can effectively improve the performance of both the policy model and the reward model.
Approach:	They propose to use Monte Carlo Tree Search for both policy model improvement and reward model improvement to bridge it to more subtle open-domain question answering.
Outcome:	The proposed approach surpasses existing methods for annotation and training data with fewer data points and achieves better performance in test-time scaling strategies.

MultiAgentBench : Evaluating the Collaboration and Competition of LLM agents (2025.acl-long)

Copied to clipboard

Kunlun Zhu, Hongyi Du, Zhaochen Hong, Xiaocheng Yang, Shuyi Guo, Zhe Wang, Zhenhailong Wang, Cheng Qian, Robert Tang, Heng Ji, Jiaxuan You

Challenge:	Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents, yet existing benchmarks focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition.
Approach:	They propose a benchmark to evaluate LLM-based multi-agent systems across diverse, interactive scenarios.
Outcome:	The proposed framework measures task completion and quality of collaboration and competition using novel, milestone-based key performance indicators.

Black-Box Tuning of Vision-Language Models with Effective Gradient Approximation (2023.findings-emnlp)

Copied to clipboard

Zixian Guo, Yuxiang Wei, Ming Liu, Zhilong Ji, Jinfeng Bai, Yiwen Guo, Wangmeng Zuo

Benchmarking Fine-Grained Error Detection in Multimodal Reasoning (2026.acl-long)

Challenge:	Large vision-language models are often not open-source due to preventing abuse or commercial factors.
Approach:	They propose a method for parameter-efficient fine-tuning to improve model accessibility . large models are often not open-source due to preventing abuse or commercial factors . they propose implementing a lightweight adapter over the output feature of an inaccessible model .
Outcome:	The proposed methods improve on 11 benchmarks and are made publicly available.

Copied to clipboard

Chi-Min Chan, Han Zhu, Chunyang Jiang, Jiaming Ji, Juntao Dai, Wei Xue, Sirui Han, Yike Guo

SAM3-I: Segment Anything with Instructions (2026.acl-long)

Challenge:	Multimodal Process Reward Models (MPRMs) have emerged as a pivotal framework for enhancing the reasoning capabilities of Multimodal Large Language Models.
Approach:	They propose a benchmark specifically designed to evaluate MPRMs’ proficiency in detecting erroneous reasoning steps across diverse error categories.
Outcome:	The proposed model achieves up to 4.8% performance improvement through test-time scaling.

Copied to clipboard

Jingjing Li, Yue Feng, Yuchen Guo, Jincai Huang, Wei Ji, Qi Bi, Yongri Piao, Miao Zhang, Xiaoqi Zhao, Qiang Chen, Shihao Zou, Huchuan Lu, Li Cheng

Low-Resource NLP

Challenge:	Existing methods for concept-level grounding and instruction-level reasoning use coarse representations and iterative mask filtering.
Approach:	They propose an instruction-following extension of the Segment Anything Model 3 family that unifies concept-level grounding and instruction-level reasoning within a single segmentation framework.
Outcome:	Experiments show that SAM3-I achieves appealing performance across referring and reasoning-based segmentation while maintaining its strong concept recall ability.

MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding (2026.acl-long)

Copied to clipboard

Fuwen Luo, Shengfeng Lou, Chi Chen, Ziyue Wang, Chenliang Li, Weizhou Shen, Jiyue Guo, Peng Li, Ming Yan, Ji Zhang, Fei Huang, Yang Liu

Visual Data in NLP

Challenge:	Existing methods for MLLMs struggle with fine-grained temporal reasoning . despite advances in video understanding, current methods struggle with time-sensitive tasks .
Approach:	They propose a time-stamp-aware multi-segment grounding method that enhances temporal understanding by introducing timestamps.
Outcome:	The proposed method outperforms existing methods on time-sensitive tasks and generalizes well across diverse temporal understanding scenarios.

Analyzing and Internalizing Complex Policy Documents for LLM Agents (2026.acl-long)

Copied to clipboard

Jiateng Liu, Zhenhailong Wang, Xiaojiang Huang, Yingjie Li, Xiang Li, Chenlei Guo, Xing Fan, Ruhi Sarikaya, Heng Ji

JurisBench: A Deep Benchmark for Assessing Large Language Models in Professional Legal Practice (2026.acl-long)

Challenge:	Large language model agents rely on in-context policy documents to act as effective user assistants.
Approach:	They propose an agentic benchmark generator with Controllable Complexity in agent policy across four levels to evaluate agents under increasing complexity.
Outcome:	The proposed method outperforms the baseline in data-sparse and high-complexity settings.

Copied to clipboard

Ziang Chen, Guannan Li, Fanlin Ji, Yipeng Kang, Jiaqi Li, Muhan Zhang, Yangtao Zhang, Li Tianjiao, Jiannan Wang, Xin Guo, Song-Chun Zhu, Bin Ling

LegalReasoner: Step-wised Verification-Correction for Legal Judgment Reasoning (2025.acl-long)

Challenge:	Existing legal benchmarks evaluate isolated tasks or exam-style questions, failing to capture the procedural interdependencies and adjudicative rigor inherent in professional practice.
Approach:	They propose a vertical, depth-oriented, domain-specific benchmark to evaluate Large Language Models (LLMs) in Chinese civil litigation.
Outcome:	The proposed benchmarks show that large language models exhibit an "illusion of competence" the results highlight a critical gap between fluent linguistic output and judicial reliability .

Copied to clipboard

Weijie Shi, Han Zhu, Jiaming Ji, Mengze Li, Jipeng Zhang, Ruiyuan Zhang, Jia Zhu, Jiajie Xu, Sirui Han, Yike Guo

Reasoning

Challenge:	Existing legal judgment prediction methods struggle with logical errors when conducting complex legal reasoning.
Approach:	They propose a method which enhances LJP reliability through step-wise verification and correction of the reasoning process.
Outcome:	The proposed model significantly improves concordance with court decisions from 72.37 to 80.27 on LLAMA-3.1-70B.

PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference (2025.acl-long)

Copied to clipboard

Jiaming Ji, Donghai Hong, Borong Zhang, Boyuan Chen, Josef Dai, Boren Zheng, Tianyi Alex Qiu, Jiayi Zhou, Kaile Wang, Boxun Li, Sirui Han, Yike Guo, Yaodong Yang

Challenge:	Using large-scale annotation data, large language models can generate noise, errors and biases, leading to unexpected behaviours.
Approach:	They propose a dataset to promote safety alignment in large language models . they separate helpfulness and harmlessness annotations for question-answering pairs .
Outcome:	The proposed dataset provides 44.6k prompts and 265k question-answer pairs with safety meta-labels for 19 harm categories and three severity levels, with answers generated by Llama-family models.

NORMSAGE: Multi-Lingual Multi-Cultural Norm Discovery from Conversations On-the-Fly (2023.emnlp-main)

Copied to clipboard

Yi Fung, Tuhin Chakrabarty, Hao Guo, Owen Rambow, Smaranda Muresan, Heng Ji

Dialogue Systems & Conversational Agents

Challenge:	Existing methods to understand acceptable behavior have focused on a single culture and manually built datasets from non-conversational settings.
Approach:	They propose a framework to automatically extract culture-specific norms from multi-lingual conversations.
Outcome:	The proposed framework extracts culture-specific norms from multi-lingual conversations.

ABC-Bench: Benchmarking Agentic Backend Coding in Real-World Development (2026.findings-acl)

Copied to clipboard

Jie Yang, Honglin Guo, Li Ji, Jiazheng Zhou, Rui Zheng, Zhikai Lei, Shuo Zhang, Zhiheng Xi, Shichun Liu, Yuxin Wang, Bo Wang, Yining Zheng, Tao Gui, Xipeng Qiu

SafeLawBench: Towards Safe Alignment of Large Language Models (2025.findings-acl)

Challenge:	Large Language Models (LLMs) have redefined the role of AI in software engineering . current benchmarks focus on localized code generation, but neglect dynamic, full-process requirements of real-world engineering.
Approach:	They propose a benchmark to evaluate agentic backend coding within a realistic, executable workflow.
Outcome:	The ABC-Bench benchmark evaluates agentic backend coding within a realistic, executable workflow.

Copied to clipboard

Chuxue Cao, Han Zhu, Jiaming Ji, Qichao Sun, Zhenghao Zhu, Wu Yinyu, Josef Dai, Yaodong Yang, Sirui Han, Yike Guo

DROWN: Towards Tighter LiRPA-based Robustness Certification (2025.coling-main)

Challenge:	Recent studies indicate that large language models (LLMs) may exhibit risks, including threats to the protection of private data and the generation of hallucinations.
Approach:	They propose to evaluate LLMs from a legal perspective using the SafeLawBench benchmark.
Outcome:	The proposed framework categorizes safety risks into three levels based on legal standards and includes 24,860 multi-choice questions and 1,106 open-domain question-answering tasks.

Copied to clipboard

Yunruo Zhang, Tianyu Du, Shouling Ji, Shanqing Guo

Robustness in NLP

Challenge:	Existing methods for certifying the robustness of deep neural networks suffer from precision or scalability issues.
Approach:	They propose a method to certify the robustness of deep neural networks . they propose to use two pairs of linear bounds to refine pre-activation bounds .
Outcome:	The proposed method achieves higher certified robustness than the baseline on CNNs and 4.68 times larger certified radii than the Transformers.

When Slower Isn’t Truer: Inverse Scaling Law of Truthfulness in Multimodal Reasoning (2026.findings-acl)

Copied to clipboard

Sitong Fang, Wenjing Cao, Jiahao Li, Xuyao Wang, Chi-Min Chan, Sirui Han, Juntao Dai, Yike Guo, Yaodong Yang, Jiaming Ji

Responsible & Trustworthy NLP

Challenge:	a study of slow reasoning models for multimodal reasoning finds that they are more prone to fabricating plausible yet false details when confronted with incomplete or misleading visual inputs.
Approach:	They conduct the first systematic study of the inverse scaling law in slow-thinking paradigms for multimodal reasoning.
Outcome:	The findings suggest that slower reasoning models are more prone to fabricating false details . the study analyzed 5,000-sample hierarchical prompt dataset by 50 participants .

Towards Economical Inference: Enabling DeepSeek’s Multi-Head Latent Attention in Any Transformer-based LLMs (2025.acl-long)

Copied to clipboard

Tao Ji, Bin Guo, Yuanbin Wu, Qipeng Guo, Shenlixing Shenlixing, Chenzhan Chenzhan, Xipeng Qiu, Qi Zhang, Tao Gui

InSerter: Speech Instruction Following with Unsupervised Interleaved Pre-training (2025.acl-long)

Challenge:	Multi-head Latent Attention (MLA) is an innovative architecture designed to ensure efficient and economical inference by significantly compressing the Key-Value (KV) cache into a latent vector.
Approach:	They propose a data-efficient fine-tuning method for transitioning from MHA to MLA using a latent vector cache.
Outcome:	The proposed architecture reduces the KV cache size of Llama2-7B by 92.19%, with only 1% drop in LongBench performance.

Copied to clipboard

Dingdong Wang, Jin Xu, Ruihang Chu, Zhifang Guo, Xiong Wang, Jincenzi Wu, Dongchao Yang, Shengpeng Ji, Junyang Lin

Isotropy-Enhanced Conditional Masked Language Models (2023.findings-emnlp)

Challenge:	Recent advances in speech large language models exhibit suboptimal performance in adhering to speech instructions.
Approach:	They propose a method to pre-train large-scale unsupervised speech-text sequences . they use text-to-speech conversion to generate textual continuations corresponding to provided speech segments .
Outcome:	The proposed model achieves superior or competitive results across diverse speech processing tasks.

Copied to clipboard

Pei Guo, Yisheng Xiao, Juntao Li, Yixin Ji, Min Zhang

Benchmarking Multi-National Value Alignment for Large Language Models (2025.findings-acl)

Challenge:	Existing non-autoregressive models with auto-regressing decoding paradigms have been used for various text generation tasks to accelerate inference but at the cost of generation quality to some extent.
Approach:	They propose to use Look Neighbors strategy to enhance learning of target token representations during training to achieve a good balance between inference speedup and generation quality.
Outcome:	The proposed models outperform current models on 4 WMT datasets and outperformed the current SoTA results.

Copied to clipboard

Chengyi Ju, Weijie Shi, Chengzhong Liu, Jiaming Ji, Jipeng Zhang, Ruiyuan Zhang, Jiajie Xu, Yaodong Yang, Sirui Han, Yike Guo

FinMME: Benchmark Dataset for Financial Multi-Modal Reasoning Evaluation (2025.acl-long)

Challenge:	Existing studies on large language models focus on ethical reviews, failing to capture the diversity of national values.
Approach:	They propose a national value extraction pipeline to efficiently construct value assessment datasets and a model-based model with instruction tagging to process raw data sources.
Outcome:	The proposed benchmark evaluates the alignment of LLMs with the values of five major nations: China, the United States, the UK, France, and Germany.

Copied to clipboard

Junyu Luo, Zhizhuo Kou, Liming Yang, Xiao Luo, Jinsheng Huang, Zhiping Xiao, Jingshu Peng, Chengzhong Liu, Jiaming Ji, Xuanzhe Liu, Sirui Han, Ming Zhang, Yike Guo