Papers by Rui Cao
AppBench: Planning of Multiple APIs from Various APPs for Complex User Instruction (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing state-of-the-art Large Language Models (LLMs) still cannot perform well in this situation even with the help of in-context learning and finetuning. |
| Approach: | They propose a benchmark to evaluate LLMs’ ability to plan and execute multiple APIs from various sources in order to complete the user’s task. |
| Outcome: | The proposed benchmarks show that the existing state-of-the-art LLMs still cannot perform well in this situation even with in-context learning and finetuning. |
LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing (2024.emnlp-main)
Copied to clipboard
Jiangshu Du, Yibo Wang, Wenting Zhao, Zhongfen Deng, Shuaiqi Liu, Renze Lou, Henry Zou, Pranav Narayanan Venkit, Nan Zhang, Mukund Srinath, Haoran Zhang, Vipul Gupta, Yinghui Li, Tao Li, Fei Wang, Qin Liu, Tianlin Liu, Pengzhi Gao, Congying Xia, Chen Xing, Cheng Jiayang, Zhaowei Wang, Ying Su, Raj Shah, Ruohao Guo, Jing Gu, Haoran Li, Kangda Wei, Zihao Wang, Lu Cheng, Surangika Ranathunga, Meng Fang, Jie Fu, Fei Liu, Ruihong Huang, Eduardo Blanco, Yixin Cao, Rui Zhang, Philip Yu, Wenpeng Yin
| Challenge: | a comparative analysis of paper (meta-)reviews by large language models (LLMs) aims to identify and distinguish LLMs from human activities . |
| Approach: | They present a comparative analysis to identify and distinguish LLM activities from human activities. |
| Outcome: | The proposed analysis aims to improve recognition of instances when someone implicitly uses LLMs for reviewing activities. |
Exploring the Impact of Negative Samples of Contrastive Learning: A Case Study of Sentence Embedding (2022.findings-acl)
Copied to clipboard
| Challenge: | Unsupervised contrastive learning is emerging as a powerful technique for extracting knowledge from unlabeled data. |
| Approach: | They propose a momentum contrastive learning model with negative sample queue for sentence embedding with a simulated model with EMA update mechanism. |
| Outcome: | The proposed model achieves a Spearman’s correlation of 77.27% on the semantic text similarity task and a maximum traceable distance metric. |
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing (2026.acl-industry)
Copied to clipboard
Junbo Niu, Zheng Liu, Zhuangcheng Gu, Bin Wang, Linke Ouyang, Zhiyuan Zhao, Tao Chu, Tianyao He, Fan Wu, Qintong Zhang, Zhenjiang Jin, Guang Liang, Rui Zhang, Wenzheng Zhang, Yuan Qu, Zhifei Ren, Yuefeng Sun, Zirui Tang, Boyu Niu, Yuanhong Zheng, Dongsheng Ma, Ziyang Miao, Hejun Dong, Siyi Qian, Junyuan Zhang, Fangdong Wang, Jingzhou Chen, Xiaomeng Zhao, Liqun Wei, Wei Li, Shasha Wang, RuiLiang Xu, Yuanyuan Cao, Lu Chen, Qianqian Wu, Huaiyu Gu, Lindong Lu, Dechen Lin, null Shenguanlin, Xuanhe Zhou, Linfeng Zhang, Yuhang Zang, Xiaoyi Dong, Jiaqi Wang, Bo Zhang, Lei Bai, Pei Chu, Weijia Li, Jiang Wu, Lijun Wu, Zhenxiang Li, Guangyu Wang, Zhongying Tu, Chao Xu, Kai Chen, Bowen Zhou, Dahua Lin, Wentao Zhang, Conghui He
| Challenge: | Document images are characterized by higher resolutions, denser content, and more complex structural layouts. |
| Approach: | They propose a 1.2B-parameter document parsing vision-language model that decouples layout analysis from local content recognition. |
| Outcome: | The proposed model surpasses general-purpose and domain-specific models on multiple benchmarks while maintaining significantly lower computational overhead. |
Robust Tool Use via Fission-GRPO: Learning to Recover from Execution Errors (2026.acl-long)
Copied to clipboard
Zhiwei Zhang, Fei Zhao, Rui Wang, Zezhong Wang, Bin Liang, Jiakang Wang, Yao Hu, Shaosheng Cao, Kam-Fai Wong
| Challenge: | Large language models (LLMs) can call tools effectively, but they remain brittle in multi-turn execution. |
| Approach: | They propose a framework that converts execution errors into on-policy corrective supervision within the RL training loop. |
| Outcome: | The proposed framework improves the error recovery rate of Qwen3-8B by 5.7% absolute and overall accuracy by 4.0% on BFCL v4 Multi-Turn. |
Multi-Programming Language Sandbox for LLMs (2025.acl-demo)
Copied to clipboard
Shihan Dou, Jiazheng Zhang, Jianxiang Zang, Yunbo Tao, Weikang Zhou, Haoxiang Jia, Shichun Liu, Yuming Yang, Shenxi Wu, Zhiheng Xi, Muling Wu, Rui Zheng, Changze Lv, Limao Xiong, Shaoqing Zhang, Lin Zhang, Wenyu Zhan, Rongxiang Weng, Jingang Wang, Xunliang Cai, Yueming Wu, Ming Wen, Yixin Cao, Tao Gui, Xipeng Qiu, Qi Zhang, Xuanjing Huang
| Challenge: | MPLSandbox is an out-of-the-box multi-programming language sandbox designed to provide unified and comprehensive feedback from compiler and analysis tools for Large Language Models (LLMs). |
| Approach: | They propose a multi-programming language sandbox that provides unified feedback from compilers and analysis tools for Large Language Models. |
| Outcome: | The proposed multi-language sandbox can provide comprehensive feedback from compilers and analysis tools for large language models (LLMs). |
Dual-Path Counterfactual Integration for Multimodal Aspect-Based Sentiment Classification (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for multimodal aspect-based sentiment classification rely on superficial correlations and spurious cues. |
| Approach: | They propose a Dual-Path Counterfactual Integration framework that explicitly models counterfactual reasoning in multimodal contexts. |
| Outcome: | The proposed framework improves model robustness by explicitly modeling counterfactual reasoning in multimodal contexts. |
Reasoning over Precedents Alongside Statutes: Case-Augmented Deliberative Alignment for LLM Safety (2026.acl-long)
Copied to clipboard
Can Jin, Rui Wu, Tong Che, Qixin Zhang, Hongwu Peng, Jiahui Zhao, Zhenting Wang, Wenqi Wei, Ligong Han, Zhao Zhang, Yuan Cao, Ruixiang Tang, Dimitris N. Metaxas
| Challenge: | OpenAI introduces deliberative alignment (DA) to enhance safety of its o-series models, but effectiveness of this approach in open-source LLMs is understudied. |
| Approach: | They propose a case-augmented deliberative alignment method for large language models . they propose to use reinforcement learning on self-generated safety reasoning chains . |
| Outcome: | The proposed method avoids narrowly enumerated rules and allows broader adaptability. |
Prompting for Multimodal Hateful Meme Classification (2022.emnlp-main)
Copied to clipboard
| Challenge: | Hateful meme classification requires complex reasoning and contextual background knowledge. |
| Approach: | They propose a simple yet effective prompt-based model that prompts pre-trained language models for hateful meme classification. |
| Outcome: | The proposed model outperforms state-of-the-art models on hateful meme classification task. |
CAPE: A Chinese Dataset for Appraisal-based Emotional Generation in Large Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing LLMs fail to capture the nuances of human emotions, making their interactions seem impersonal or inadequate. |
| Approach: | They propose a two-stage automatic data generation framework to generate a Chinese dataset called CAPE . their data is a cognitive appraisal theory-based Emotional corpus that accounts for personal and situational factors. |
| Outcome: | The proposed framework can generate human-like responses in conversation with large language models. |
Evaluating LLMs’ Assessment of Mixed-Context Hallucination Through the Lens of Summarization (2025.findings-acl)
Copied to clipboard
| Challenge: | Large language models generate coherent text and follow instructions across diverse tasks, but a critical challenge in scaling LLM applications is hallucination, where the generated content lacks factual grounding or deviates from the intended discourse context. |
| Approach: | They use summarization as a representative task to evaluate LLMs' capability in detecting mixed-context hallucinations, specifically distinguishing between factual and non-factual hallucinos. |
| Outcome: | The proposed model distinguishes between factual and non-factual hallucinations, and their performance bottlenecks. |
Modularized Zero-shot VQA with Pre-trained Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent work on zero-shot visual question answering does not explicitly consider multi-step reasoning chains, making them less interpretable compared with a decomposition-based approach. |
| Approach: | They propose a modularized zero-shot network that explicitly decomposes questions into sub reasoning steps and is highly interpretable. |
| Outcome: | The proposed model decomposes questions into sub reasoning steps and is highly interpretable. |
Curr-ReFT: Overcoming Training Bottlenecks in Small-scale Vision-Language Models via Curriculum Reinforcement Finetuning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | State-of-the-art vision-language models require massive scaling that limits practical deployment. |
| Approach: | They propose to use supervised fine-tuning to train small-scale vision-language models but face out-of-domain collapse when trained with traditional supervised learning (SFT). |
| Outcome: | Experiments show that curr-reFT achieves state-of-the-art performance across visual tasks in both in- and out-of domain settings and benchmarks. |
CCTVBench: Contrastive Consistency Traffic VideoQA Benchmark for Multimodal LLMs (2026.findings-acl)
Copied to clipboard
Xingcheng Zhou, Hao Guo, Rui Song, Walter Zimmer, Mingyu Liu, André Schamschurko, Hu Cao, Alois Knoll
| Challenge: | Existing Vision-language models are prone to hallucinating nonexistent entities or events and missing subtle but critical visual cues. |
| Approach: | They propose a Traffic VideoQA Benchmark that enforces a single structured decision pattern over each video question quadruple and provides actionable diagnostics that decompose failures into positive omission, positive swap, negative hallucination, mutual-exclusivity violation. |
| Outcome: | The proposed model detects true hazards when an accident occurs, and rejects plausible-but-false hypotheses under near-identical counterfactual scenes. |
Knowledge Generation for Zero-shot Knowledge-based VQA (2024.findings-eacl)
Copied to clipboard
| Challenge: | Recent knowledge-based visual question answering methods do not explicitly show the knowledge needed to answer the questions and therefore lack interpretability. |
| Approach: | They propose a method which generates knowledge from an LLM and incorporates it into a zero-shot manner. |
| Outcome: | The proposed method performs better than previous zero-shot K-VQA methods on two benchmarks and is generally relevant and helpful. |
From Trajectories to Graphs: Contract-Checked Editing for Verifier-Guided LLM Reasoning (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for inference-time search refine single trajectories and lack a reliable mechanism for composing partial solutions across candidates. |
| Approach: | a new method uses a gate-based algorithm to validate a nontrivial edit before invoking the verifier. |
| Outcome: | a new method improves verifier-runnable recombination and accuracy over existing methods . it outperforms execution-guided beam search on Spider and humanEval-MF on MCTS . a contract-checked graph editing improves recompilation and recombines partial solutions . |
Aspect Is Not You Need: No-aspect Differential Sentiment Framework for Aspect-based Sentiment Analysis (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing approaches to classify aspects with aspect sentiment bias are hard to find . |
| Approach: | They propose a no-aspect differential sentiment framework for the ABSA task that eliminates aspect sentiment bias and uses differential sentiment loss instead of cross-entropy loss to better classify the sentiments. |
| Outcome: | The proposed framework can be combined with almost all traditional ABSA methods. |
Entity Relation Extraction as Dependency Parsing in Visually Rich Documents (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on key information extraction from visually rich documents focus on labeling the text within bounding boxes, while relations between words are unexplored. |
| Approach: | They propose to use a dependency parsing model to extract semantic entities from visually rich documents by combining entity labeling and relation extraction tasks. |
| Outcome: | The proposed model achieves 65.96% F1 score on the FUNSD dataset. |
HateGAN: Adversarial Generative-Based Data Augmentation for Hate Speech Detection (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods to detect online hate speech depend heavily on labeled datasets for training, which results in poor detection performance of the hate speech class. |
| Approach: | They propose a deep generative reinforcement learning model which augments two commonly-used hate speech detection datasets with the HateGAN generated tweets. |
| Outcome: | The proposed model improves the detection performance of hate speech class regardless of the classifiers and datasets used in the detection task. |
SSP: Self-Supervised Post-training for Conversational Search (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to search for information are limited by data scarcity and lack contextual semantic understanding. |
| Approach: | They propose a new post-training paradigm with three self-supervised tasks to initialize the conversational search model to enhance dialogue structure and contextual semantic understanding. |
| Outcome: | The proposed method can initialize the conversational search model to enhance dialogue structure and contextual semantic understanding. |
Recent Advances in Online Hate Speech Moderation: Multimodality and the Role of Large Models (2024.findings-emnlp)
Copied to clipboard
| Challenge: | HS is any communication demeaning a person or a group based on social or ethnic characteristics that undermines social harmony and individual safety . the recent Israel-Hamas conflict has escalated both anti-Muslim and anti-Semitic sentiments worldwide . |
| Approach: | They examine the role of large language models and large multimodal models in HS moderation . they examine how text, images, and audio interact to spread hate speech . |
| Outcome: | The findings highlight the need for solutions in low-resource settings and highlight the gaps in existing methods. |