Papers by Haozhe Chen
A Goal Without a Plan Is Just a Wish: Efficient and Effective Global Planner Training for Long-Horizon Agent Tasks (2026.acl-long)
Copied to clipboard
Shuzheng Si, Haozhe Zhao, Kangyang Luo, Gang Chen, Fanchao Qi, Minjia Zhang, Baobao Chang, Maosong Sun
| Challenge: | Recent advances in large language models (LLMs) have leapt from static chatbots to versatile agents that tackle complex tasks such as science experiments. |
| Approach: | They propose a plan-and-execute framework and propose 'EAGLET' to enhance the executor agent's planning abilities without human effort. |
| Outcome: | The proposed method outperforms existing methods on three long-horizon tasks and reduces training costs by 8 compared to baselines. |
FaithLens: Detecting and Explaining Faithfulness Hallucination (2026.findings-acl)
Copied to clipboard
Shuzheng Si, Qingyi Wang, Haozhe Zhao, Yuzhuo Bai, Guanqiao Chen, Kangyang Luo, Gang Chen, Fanchao Qi, Minjia Zhang, Baobao Chang, Maosong Sun
| Challenge: | Recent progress in large language models (LLMs) has revolutionized text generation. |
| Approach: | They propose a faithfulness hallucination detection model that can provide binary predictions and corresponding explanations to improve trustworthiness. |
| Outcome: | The proposed model outperforms advanced models on 12 diverse tasks. |
Robustness via Referencing: Defending against Prompt Injection Attacks by Referencing the Executed Instruction (2026.findings-acl)
Copied to clipboard
Yulin Chen, Haoran Li, Yuan Sui, Yue Liu, Yufei He, Xiaoling Bai, Chi Fei, Li Yabo, Haozhe Ma, Yangqiu Song, Bryan Hooi
| Challenge: | Prompt injection attacks manipulate large language models (LLMs) by misleading them to deviate from the original input instructions and execute maliciously injected instructions. |
| Approach: | They propose a prompt injection defense method that suppresses the model's instruction-following tendencies rather than suppressing them. |
| Outcome: | The proposed method outperforms prompt-engineering-based approaches and fine-tuning methods and reduces the ASR to nearly 0% in some scenarios. |
Infinity-Parser: Layout-Aware Reinforcement Learning with High-quality Document Parsing Dataset (2026.findings-acl)
Copied to clipboard
Baode Wang, Biao Wu, Weizhen Li, Meng Fang, Zuming Huang, Jun Huang, Yanjie Liang, Haozhe Wang, Ling Chen, Wei Chu, Yuan Qi
| Challenge: | Existing supervised fine-tuning methods struggle to generalize across document types, leading to poor performance. |
| Approach: | They propose layoutRL, a reinforcement learning framework that optimizes layout understanding through composite rewards integrating normalized edit distance, paragraph count accuracy, and reading order preservation. |
| Outcome: | The proposed model outperforms specialized document parsing systems and general-purpose vision-language models on a broad range of document types, languages, and structural complexities. |
Self-Distillation Bridges Distribution Gap in Language Model Fine-Tuning (2024.acl-long)
Copied to clipboard
| Challenge: | Experimental results show that fine-tuning of large language models for specific tasks can be challenging . distribution shift during fine-timing can lead to performance degradation in general task capabilities . |
| Approach: | They propose a new approach that bridges the distribution gap between task datasets and LLMs by guiding fine-tuning with a distilled dataset generated by the model itself. |
| Outcome: | The proposed approach achieves comparable or superior performance on downstream tasks compared to the vanilla approach. |
EmoKnob: Enhance Voice Cloning with Fine-Grained Emotion Control (2024.emnlp-main)
Copied to clipboard
| Challenge: | EmoKnob framework allows fine-grained emotion control in speech synthesis with few-shot demonstrative samples of arbitrary emotion. |
| Approach: | They propose a framework that allows fine-grained emotion control in speech synthesis . they propose two methods to apply emotion control on emotions described by open-ended text . |
| Outcome: | The proposed framework allows fine-grained emotion control in speech synthesis with few-shot demonstrative samples of arbitrary emotion. |
Looking Beyond Text: Reducing Language Bias in Large Vision-Language Models via Multimodal Dual-Attention and Soft-Image Guidance (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large vision-language models (LVLMs) have been criticized for their language bias. |
| Approach: | They propose to use a dual-attention mechanism to construct separate attention for visual and text inputs to enhance integration of visual inputs across models. |
| Outcome: | Experiments show that the proposed model debiases LVLMs from their language bias, enhancing visual comprehension and reducing hallucinations without additional resources. |
MENTOR: Efficient Autoregressive Image Generation with Balanced Multimodal Control (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent text-to-image models achieve impressive visual quality but still face challenges in precise controllability, balancing multimodal inputs, and high training cost for multimodal image generation. |
| Approach: | They propose an autoregressive framework with a two-stage training paradigm for controllable multimodal image generation. |
| Outcome: | Extensive experiments on DreamBench++ and DreamBech show that the proposed framework achieves a strong balance between textual and visual guidance for controllable image generation. |
ACECODER: Acing Coder RL via Automated Test-Case Synthesis (2025.acl-long)
Copied to clipboard
| Challenge: | Recent coder models have been driven by supervised fine-tuning, but the potential of reinforcement learning remains unexplored due to the lack of reliable reward data/model in the code domain. |
| Approach: | They propose a pipeline that generates extensive test-case pairs from existing code data and constructs preference pairs based on pass rates over sampled programs. |
| Outcome: | The proposed pipeline generates extensive (question, test-cases) pairs from existing code data and trains them with Bradley-Terry loss. |
SWE-QA-Pro: A Representative Benchmark and Scalable Training Recipe for Repository-Level Code Understanding (2026.findings-acl)
Copied to clipboard
Songcheng Cai, Zhiheng Lyu, Yuansheng Ni, Xiangchao Chen, Baichuan Zhou, Shenzhe Zhu, Yi Lu, Haozhe Wang, Chi Ruan, Benjamin Schneider, Weixu Zhang, Xiang Li, Andy Zheng, Yuyu Zhang, Ping Nie, Wenhu Chen
| Challenge: | Existing benchmarks for agentic repository-level code understanding overlook long tail topics and rely on memorized knowledge. |
| Approach: | They propose a repository-level agentic code understanding benchmark that uses long-tail repositories with executable environments to enforce topical balance. |
| Outcome: | Empirically, a Qwen3-8B model trained with the proposed benchmark outperforms GPT-4o by 2.3 points. |
Aligning Large Language Models to Follow Instructions and Hallucinate Less via Effective Data Filtering (2025.acl-long)
Copied to clipboard
Shuzheng Si, Haozhe Zhao, Gang Chen, Cheng Gao, Yuzhuo Bai, Zhitong Wang, Kaikai An, Kangyang Luo, Chen Qian, Fanchao Qi, Baobao Chang, Maosong Sun
| Challenge: | Existing studies show that training LLMs on data containing unfamiliar knowledge during instruction tuning can encourage hallucinations. |
| Approach: | They propose a framework that measures how familiar the LLM is with instruction data and introduce an expert-aligned reward model to ensure the quality of selected samples. |
| Outcome: | The proposed framework reduces hallucinations while maintaining a competitive ability to follow instructions. |
Mitigating Language-Level Performance Disparity in mPLMs via Teacher Language Selection and Cross-lingual Self-Distillation (2024.naacl-long)
Copied to clipboard
| Challenge: | Large-scale multilingual pretrained language models (mPLMs) yield impressive performance on cross-language tasks, yet significant performance disparities exist across different languages within the same mPLm. |
| Approach: | They propose to leverage the learned knowledge from well-performing languages to guide under-performing ones within the same mPLM. |
| Outcome: | The proposed model shows that it can guide under-performing languages while minimizing language-level performance disparities across different mPLMs. |
PCA-Bench: Evaluating Multimodal Large Language Models in Perception-Cognition-Action Chain (2024.findings-acl)
Copied to clipboard
Liang Chen, Yichi Zhang, Shuhuai Ren, Haozhe Zhao, Zefan Cai, Yuchi Wang, Peiyi Wang, Xiangdi Meng, Tianyu Liu, Baobao Chang
| Challenge: | a new multimodal decision-making benchmark evaluates the integrated capabilities of multimodal large language models. |
| Approach: | They propose a multimodal decision-making benchmark for evaluating MLLMs . they propose an automatic evaluation protocol to assess 10 prevalent ML models . |
| Outcome: | The proposed benchmark improves performance of multimodal large language models in three scenarios . the model is required to integrate multiple capabilities to make accurate decisions . |
Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data? (2025.findings-acl)
Copied to clipboard
Che Liu, Zhongwei Wan, Haozhe Wang, Yinda Chen, Talha Qaiser, Chen Jin, Nikolay Burlutskiy, Fariba Yousefi, Rossella Arcucci
| Challenge: | Medical Vision-Language Pretraining (MedVLP) models typically require large-scale datasets with paired, high-quality image-text data. |
| Approach: | They propose to generate large-scale synthetic image-text pairs using off-the-shelf generative models . they propose to isolate model and training settings, focusing entirely from the data perspective. |
| Outcome: | The proposed pipeline outperforms models trained on real data by 3.8% on averaged AUC on zero-shot classification tasks. |
GATEAU: Selecting Influential Samples for Long Context Alignment (2025.emnlp-main)
Copied to clipboard
Shuzheng Si, Haozhe Zhao, Gang Chen, Yunshui Li, Kangyang Luo, Chuancheng Lv, Kaikai An, Fanchao Qi, Baobao Chang, Maosong Sun
| Challenge: | Existing studies have attempted to scale up the available data volume by synthesizing long instruction-following samples, but a lack of a well-defined strategy for ensuring data quality may introduce low-quality samples and restrict the model’s performance. |
| Approach: | They propose a framework to identify influential samples enriched with long-range dependency relations that can be used to align large language models to handle instructions with extremely long contexts. |
| Outcome: | The proposed framework identifies samples with long-range dependency relations and shows that the model trained on these samples exhibits better instruction-following and long-context understanding capabilities. |
MMEvalPro: Calibrating Multimodal Benchmarks Towards Trustworthy and Efficient Evaluation (2025.naacl-long)
Copied to clipboard
Jinsheng Huang, Liang Chen, Taian Guo, Fu Zeng, Yusheng Zhao, Bohan Wu, Ye Yuan, Haozhe Zhao, Zhihui Guo, Yichi Zhang, Jingyang Yuan, Wei Ju, Luchen Liu, Tianyu Liu, Baobao Chang, Ming Zhang
| Challenge: | Large Multimodal Models (LMMs) exhibit impressive cross-modal understanding and reasoning abilities, but many benchmarks suffer from systematic biases. |
| Approach: | They propose a benchmark to avoid Type-I errors by creating one perception question and one knowledge anchor question through a meticulous annotation process. |
| Outcome: | The proposed benchmark avoids Type-I errors while maintaining reliability of MCQ evaluations. |
Embedding and Gradient Say Wrong: A White-Box Method for Hallucination Detection (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for hallucination detection have attracted more attention from the community. |
| Approach: | They propose to model the distributional distance between the regular conditional output and the unconditional output, which is generated without a given input text. |
| Outcome: | The proposed model achieves state-of-the-art on the hallucination benchmarks HADES and other datasets. |