Papers by Renjie Liu
Progra: Progress-Aware Reinforcement Learning for Multi-Turn Function Calling (2026.findings-acl)
Copied to clipboard
Huacan Chai, Zijie Cao, Maolin Ran, Yingxuan Yang, Jianghao Lin, Xin Peng, Hairui Wang, Renjie Ding, Ziyu Wan, Muning Wen, Weiwen Liu, Weinan Zhang, Fei Huang, Ying Wen
| Challenge: | Existing methods for multi-turn function calling are limited by redundancy and lack explicit integration of progress awareness into training. |
| Approach: | They propose a framework that explicitly integrates progress awareness into LLM training for multi-turn function calling. |
| Outcome: | Empirical results show that Progra outperforms existing methods on two public benchmarks. |
Course-Correction: Safety Alignment Using Synthetic Preferences (2024.emnlp-industry)
Copied to clipboard
Rongwu Xu, Yishuo Cai, Zhenhong Zhou, Renjie Gu, Haiqin Weng, Liu Yan, Tianwei Zhang, Wei Xu, Han Qiu
| Challenge: | Recent studies show that large language models generate harmful content, but the potential for generating harmful content is an escalating concern. |
| Approach: | They propose to fine-tune LLMs with preference learning to emphasize the preference for timely course-correction by using an automated pipeline. |
| Outcome: | The proposed model improves course-correction skills without affecting general performance and resists jailbreak attacks. |
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models (2025.findings-naacl)
Copied to clipboard
Weizhe Chen, Zhicheng Zhang, Guanlin Liu, Renjie Zheng, Wenlei Shi, Chen Dun, Zheng Wu, Xing Jin, Lin Yan
| Challenge: | Large language models (LLMs) have demonstrated remarkable capabilities across various domains since the release of ChatGPT . a key challenge in developing these general capabilities is efficiently sourcing diverse, high-quality data. |
| Approach: | They introduce Flaming-hot Initiation with Regular Execution (FIRE) sampling to efficiently find good responses by promoting diversity. |
| Outcome: | The proposed method enhances inference-time generation quality and benefits training in the alignment stage. |
Mitigating the Alignment Tax of RLHF (2024.emnlp-main)
Copied to clipboard
Yong Lin, Hangyu Lin, Wei Xiong, Shizhe Diao, Jianmeng Liu, Jipeng Zhang, Rui Pan, Haoxiang Wang, Wenbin Hu, Hanning Zhang, Hanze Dong, Renjie Pi, Han Zhao, Nan Jiang, Heng Ji, Yuan Yao, Tong Zhang
| Challenge: | Large Language Models (LLMs) acquire a wide range of abilities during pre-training, but aligning LLMs under Reinforcement Learning with Human Feedback (RLHF) can lead to forgetting pretrained abilities, which is also known as the alignment tax. |
| Approach: | They propose to use a model averaging technique to find the most powerful alignment-forging Pareto front among RLHF algorithms. |
| Outcome: | The proposed method achieves the strongest alignment-forging Pareto front among competing methods. |
SceMQA: A Scientific College Entrance Level Multimodal Question Answering Benchmark (2024.acl-short)
Copied to clipboard
Zhenwen Liang, Kehan Guo, Gang Liu, Taicheng Guo, Yujun Zhou, Tianyu Yang, Jiajun Jiao, Renjie Pi, Jipeng Zhang, Xiangliang Zhang
| Challenge: | SceMQA focuses on core science subjects including Mathematics, Physics, Chemistry, and Biology. |
| Approach: | They propose to use SceMQA to evaluate multimodal question answering at college entrance level. |
| Outcome: | The proposed model provides specific knowledge points for each problem and detailed explanations for each answer. |
LongVideoAgent: Multi-Agent Reasoning with Long Videos (2026.acl-long)
Copied to clipboard
| Challenge: | a key emerging challenge is robust long video understanding, authors say . current methods compress content into lossy summaries or rely on limited toolsets . |
| Approach: | They propose a multi-agent framework where a master LLM coordinates a grounding agent and a vision agent to extract targeted textual observations. |
| Outcome: | The proposed model outperforms strong non-agent baselines on episode-level datasets . the proposed model significantly outperformed existing models on other datasets. |
STACL: Simultaneous Translation with Implicit Anticipation and Controllable Latency using Prefix-to-Prefix Framework (P19-1)
Copied to clipboard
Mingbo Ma, Liang Huang, Hao Xiong, Renjie Zheng, Kaibo Liu, Baigong Zheng, Chuanqiang Zhang, Zhongjun He, Hairong Liu, Xing Li, Hua Wu, Haifeng Wang
| Challenge: | Simultaneous translation is notoriously dif- ficult due to word-order differences. |
| Approach: | They propose a prefix-to-prefix framework that implicitly learns to anticipate in a single translation model. |
| Outcome: | The proposed framework achieves low latency and reasonable qual- ity on 4 directions. |
Simultaneous Translation Policies: From Fixed to Adaptive (2020.acl-main)
Copied to clipboard
| Challenge: | Adaptive policies can balance translation quality and latency based on context information . previous methods on obtaining adaptive policies rely on complicated training process . |
| Approach: | They propose to obtain adaptive policies by a simple heuristic composition of fixed policies . they propose to use a heurism to obtain policies that can outperform fixed ones . |
| Outcome: | Experiments on Chinese -> English and German -> english show that adaptive policies outperform fixed policies by up to 4 BLEU points for the same latency. |
Opportunistic Decoding with Timely Correction for Simultaneous Translation (2020.acl-main)
Copied to clipboard
| Challenge: | Existing approaches to balancing translation quality and latency are either too aggressive or too conservative. |
| Approach: | They propose an opportunistic decoding technique that always (over-)generates a certain mount of extra words at each step to keep the audience on track with the latest information. |
| Outcome: | The proposed technique reduces latency and increases BLEU with no over-generating . it also corrects mistakes in the overgenerated words when observing more context . |
Fluent and Low-latency Simultaneous Speech-to-Speech Translation with Self-adaptive Training (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Current approaches to simultaneous speech-to-speech translation accumulate more and more latencies in later sentences when the speaker talks faster. |
| Approach: | They propose a method which generates more fluent target speech latency than the baseline . they propose to use self-adaptive translation to adjust the length of translations to accommodate different source speech rates. |
| Outcome: | Xiong et al., 2019) show that the proposed method generates more fluent target speech latency than baseline . authors say it provides more natural communication process than speech-to-text translation . xiong and colleagues say the proposed technique is more efficient than current approaches . |
Plum: Prompt Learning using Metaheuristics (2024.findings-acl)
Copied to clipboard
Rui Pan, Shuo Xing, Shizhe Diao, Wenhe Sun, Xiang Liu, KaShun Shum, Jipeng Zhang, Renjie Pi, Tong Zhang
| Challenge: | Recent advances in prompt learning have led to a need for general prompt optimization methods. |
| Approach: | They propose a branch of discrete non-convex optimization methods with over 100 options as a promising approach to prompt learning. |
| Outcome: | The proposed methods can be used to discover more human-understandable prompts that were previously unknown in reasoning and image generation tasks. |
Attention and Edge-Label Guided Graph Convolutional Networks for Named Entity Recognition (2022.emnlp-main)
Copied to clipboard
| Challenge: | Named entity recognition (NER) is the recognition of entities with specific meanings in the text, mainly including person, organization, location, etc. |
| Approach: | They propose an edge-aware node joint update module and introduce a node-awful edge update module to explore hidden in structured information and solve the wrong dependency label information to some extent. |
| Outcome: | The proposed model can exploit the structured information on the dependency tree to improve the recognition of long entities. |
Towards an On-device Agent for Text Rewriting (2024.findings-naacl)
Copied to clipboard
Yun Zhu, Yinxiao Liu, Felix Stahlberg, Shankar Kumar, Yu-Hui Chen, Liangchen Luo, Lei Shu, Renjie Liu, Jindong Chen, Lei Meng
| Challenge: | Large Language Models (LLMs) have demonstrated impressive capabilities for text rewriting, however creating a smaller yet potent language model presents two formidable challenges: costly data collection and absence of emergent capabilities. |
| Approach: | They propose a new instruction tuning method to develop a mo-bile text rewriting model that leverages LLM-generated data and heuristic reinforcement learning, eliminating the need for human data collection. |
| Outcome: | The proposed model surpasses the current state-of-the-art LLMs in text rewriting while maintaining a significantly reduced model size using public benchmark EditEval and our new benchmark. |
MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices (2020.acl-main)
Copied to clipboard
| Challenge: | Empirical studies show that MobileBERT is 4.3x smaller and 5.5x faster than BERT_BASE . BERT is one of the largest models ever in NLP, but suffers from heavy model size and high latency . |
| Approach: | They propose a tool to compress and accelerate the popular BERT model by task-agnostic application. |
| Outcome: | The proposed model is 4.3x smaller and 5.5x faster than BERT_BASE . it achieves competitive results on well-known benchmarks . |
Knowledge Extraction on Semi-Structured Content: Does It Remain Relevant for Question Answering in the Era of LLMs? (2026.eacl-long)
Copied to clipboard
Kai Sun, Yin Huang, Srishti Mehra, Mohammad Kachuee, Xilun Chen, Renjie Tao, Zhaojiang Lin, Andrea Jessee, Nirav Shah, Alex L Betty, Yue Liu, Anuj Kumar, Wen-tau Yih, Xin Luna Dong
| Challenge: | Existing literature on knowledge extraction for question answering questions whether it is still relevant for question answerrs. |
| Approach: | They extend an existing benchmark with knowledge extraction annotations and evaluate commercial and open-source LLMs of varying sizes. |
| Outcome: | The proposed model can achieve high QA accuracy, but can still benefit from knowledge extraction through augmentation with extracted triples and multi-task learning. |
MR. Judge: Multimodal Reasoner as a Judge (2025.emnlp-main)
Copied to clipboard
| Challenge: | Effective reward modeling is especially valuable in reinforcement learning (RLHF) . |
| Approach: | They propose a paradigm for empowering general-purpose MLLMs judges with strong reasoning capabilities by using multiple-choice problem models instead of directly assigning scores. |
| Outcome: | The proposed model surpasses GPT-4o on VL-RewardBench and improves performance on MM-Vet by up to 7.7%. |
Pointing to a Llama and Call it a Camel: On the Sycophancy of Multimodal Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Multimodal large language models exhibit a pronounced form of visual sycophantic behavior when they process image inputs. |
| Approach: | They propose a technique that allows multimodal large language models to engage in reflective reasoning and determine whether a user’s instruction is misleading or corrective. |
| Outcome: | The proposed model resists misleading instructions but is stubborn even if it is wrong. |
Bridge-Coder: Transferring Model Capabilities from High-Resource to Low-Resource Programming Language (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) excel at generating code for high-resource programming languages (HRPLs) however, they struggle significantly with low-resourced programming languages such as D, exacerbating the digital divide. |
| Approach: | They propose a method to generate LRPL data using LLM's general knowledge, HRPL proficiency, and in-context learning capabilities. |
| Outcome: | The proposed method improves on R, D, Racket, and Bash, while maintaining the same quality. |
OlympiadBench: A Challenging Benchmark for Promoting AGI with Olympiad-Level Bilingual Multimodal Scientific Problems (2024.acl-long)
Copied to clipboard
Chaoqun He, Renjie Luo, Yuzhuo Bai, Shengding Hu, Zhen Thai, Junhao Shen, Jinyi Hu, Xu Han, Yujie Huang, Yuxiang Zhang, Jie Liu, Lei Qi, Zhiyuan Liu, Maosong Sun
| Challenge: | Large Language Models (LLMs) and Large Multimodal Models have exceeded general human capabilities in various tasks. |
| Approach: | They present an Olympiad-level bilingual multimodal scientific benchmark featuring 8,476 problems from Olympiad level mathematics and physics competitions. |
| Outcome: | The best performing model, GPT-4V, attains an average score of 17.97% on OlympiadBench, with a mere 10.74% in physics, highlighting the benchmark rigor and the intricacy of physical reasoning. |
OpenResearcher: Unleashing AI for Accelerated Scientific Research (2024.emnlp-demo)
Copied to clipboard
Yuxiang Zheng, Shichao Sun, Lin Qiu, Dongyu Ru, Cheng Jiayang, Xuefeng Li, Jifan Lin, Binjie Wang, Yun Luo, Renjie Pan, Yang Xu, Qingkai Min, Zizhao Zhang, Yiwen Wang, Wenjie Li, Pengfei Liu
| Challenge: | Global scientific publications are growing annually by about 4%-5% (Pinedo et al., 2024). |
| Approach: | They introduce an AI-assisted platform that answers diverse questions from researchers using Retrieval-Augmented Generation (RAG) they develop various tools to understand queries, search from the scientific literature, filter retrieved information, provide accurate and comprehensive answers, and self-refine answers. |
| Outcome: | OpenResearcher is built on Retrieval-Augmented Generation (RAG) to integrate Large Language Models (LLMs) with up-to-date, domain-specific knowledge. |
Unlearners Can Lie: Evaluating and Improving Honesty in LLM Unlearning (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for unlearning in large language models often hallucinate, generate abnormal token sequences, or behave inconsistently, raising safety and trust concerns. |
| Approach: | They propose a formal definition of unlearning honesty that preserves both utility and honesty on retained knowledge and ensures effective forgetting while encouraging the model to acknowledge its limitations. |
| Outcome: | The proposed method achieves highest rejection rate and refusal stability on Q A tasks from the forget set, nearly double the second-best method. |
UltraEval: A Lightweight Platform for Flexible and Comprehensive Evaluation for LLMs (2024.acl-demos)
Copied to clipboard
Chaoqun He, Renjie Luo, Shengding Hu, Ranchi Zhao, Jie Zhou, Hanghao Wu, Jiajie Zhang, Xu Han, Zhiyuan Liu, Maosong Sun
| Challenge: | Existing evaluation platforms are complex and poorly modularized, hindering seamless incorporation into researcher’s workflows. |
| Approach: | They propose a lightweight evaluation framework characterized by lightweight, comprehensiveness, modularity, and efficiency that integrates models, data, and metrics into a unified evaluation workflow. |
| Outcome: | The proposed evaluation framework is lightweight, comprehensive, modular, and efficient. |
Incremental Text-to-Speech Synthesis with Prefix-to-Prefix Framework (2020.findings-emnlp)
Copied to clipboard
Mingbo Ma, Baigong Zheng, Kaibo Liu, Renjie Zheng, Hairong Liu, Kainan Peng, Kenneth Church, Liang Huang
| Challenge: | Text-to-speech synthesis (TTS) has seen rapid progress in recent years, but still suffers from latencies. |
| Approach: | They propose a neural incremental TTS approach that synthesizes speech in an online fashion, playing a segment of audio while generating the next. |
| Outcome: | Experiments on English and Chinese TTS show that the proposed approach achieves similar speech naturalness compared to full sentence TTS, but with a constant (1-2 words) latency. |
Proofread: Fixes All Errors with One Tap (2024.acl-demos)
Copied to clipboard
Renjie Liu, Yanxiang Zhang, Yun Zhu, Haicheng Sun, Yuanbo Zhang, Michael Huang, Shanqing Cai, Lei Meng, Shumin Zhai
| Challenge: | Extensive experiments on a human-labeled golden set showed our tuned PaLM2-XS model achieved 85.56% good ratio. |
| Approach: | They propose a two-stage tuning approach to acquire the dedicated Large Language Model for the feature, followed by a reinforcement learning approach for targeted refinement. |
| Outcome: | The proposed model achieves 85.56% good quality on Rewrite and proofread tasks on human-labeled golden sets. |