Papers by James Liu
SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning (2024.emnlp-main)
Copied to clipboard
Jinghan Jia, Yihua Zhang, Yimeng Zhang, Jiancheng Liu, Bharat Runwal, James Diffenderfer, Bhavya Kailkhura, Sijia Liu
| Challenge: | Large Language Models (LLMs) have highlighted the need for effective unlearning mechanisms to comply with data regulations and ethical AI practices. |
| Approach: | They propose a second-order optimization-based LLM unlearning framework which extends the static, one-shot model update using influence unlearning to a dynamic, iterative unlearning process. |
| Outcome: | The proposed framework outperforms first-order methods across unlearning tasks, models, and metrics. |
FOLIO: Natural Language Reasoning with First-Order Logic (2024.emnlp-main)
Copied to clipboard
Simeng Han, Hailey Schoelkopf, Yilun Zhao, Zhenting Qi, Martin Riddell, Wenfei Zhou, James Coady, David Peng, Yujie Qiao, Luke Benson, Lucy Sun, Alexander Wardle-Solano, Hannah Szabó, Ekaterina Zubova, Matthew Burtell, Jonathan Fan, Yixin Liu, Brian Wong, Malcolm Sailor, Ansong Ni, Linyong Nan, Jungo Kasai, Tao Yu, Rui Zhang, Alexander Fabbri, Wojciech Kryscinski, Semih Yavuz, Ye Liu, Xi Lin, Shafiq Joty, Yingbo Zhou, Caiming Xiong, Rex Ying, Arman Cohan, Dragomir Radev
| Challenge: | Existing benchmarks for logical reasoning in large language models lack language naturalness or limited complexity. |
| Approach: | They propose to use first-order logic annotations to evaluate logical reasoning capabilities of large language models. |
| Outcome: | The proposed dataset evaluates the FOL reasoning ability of supervised fine-tuning on medium-sized language models. |
DisastIR: A Comprehensive Information Retrieval Benchmark for Disaster Management (2025.findings-emnlp)
Copied to clipboard
Kai Yin, Xiangjue Dong, Chengkai Liu, Lipai Huang, Yiming Xiao, Zhewei Liu, Ali Mostafavi, James Caverlee
| Challenge: | Existing information retrieval benchmarks focus on general or specialized domains, such as medicine or finance, neglecting the unique linguistic complexity and diverse information needs encountered in disaster management scenarios. |
| Approach: | DisastIR is the first comprehensive IR evaluation benchmark specifically tailored for disaster management. |
| Outcome: | DisastIR covers 48 retrieval tasks derived from six search intents and eight general disaster categories . evaluations show no single model excelling universally . |
TextGenSHAP: Scalable Post-Hoc Explanations in Text Generation with Long Documents (2024.findings-acl)
Copied to clipboard
| Challenge: | Large language models (LLMs) are difficult to explain and understand due to long input contexts and autoregressive output generation. |
| Approach: | They propose a post-hoc explanation method which incorporates LLM-specific techniques. |
| Outcome: | The proposed method improves retrieval recall and prediction accuracy significantly on open-domain question answering benchmarks. |
A Survey on LLMs for Story Generation (2025.findings-emnlp)
Copied to clipboard
Maria Teleki, Vedangi Bengali, Xiangjue Dong, Sai Tejas Janjur, Haoran Liu, Tian Liu, Cong Wang, Ting Liu, Yin Zhang, Frank Shipman, James Caverlee
| Challenge: | Methods for story generation with Large Language Models (LLMs) have come into the spotlight recently. |
| Approach: | They propose a novel taxonomy of LLMs for story generation consisting of two major paradigms: independent story generation by an LLM, and author-assistance for story creation . |
| Outcome: | The proposed taxonomy compares existing work on the topic with those of novel author-assistance models. |
InstructCoder: Instruction Tuning Large Language Models for Code Editing (2024.acl-srw)
Copied to clipboard
| Challenge: | InstructCoder is the first instruction-tuning dataset designed to adapt LLMs for general-purpose code editing. |
| Approach: | They propose to use Large Language Models to edit code based on user instructions . they use a dataset to adapt LLMs to general-purpose code editing . |
| Outcome: | The proposed model can significantly improve code editing performance compared to proprietary models . the proposed model is based on a human-written execution-based benchmark . |
Werewolf Among Us: Multimodal Resources for Modeling Persuasion Behaviors in Social Deduction Games (2023.findings-acl)
Copied to clipboard
Bolin Lai, Hongxin Zhang, Miao Liu, Aryan Pariani, Fiona Ryan, Wenqi Jia, Shirley Anugrah Hayati, James Rehg, Diyi Yang
| Challenge: | Existing studies on persuasive behavior modeling focus on textual dialogues . a multimodal dataset is available for persuasion modeling . |
| Approach: | They propose a multimodal dataset for modeling persuasive behaviors using visual signals. |
| Outcome: | The proposed dataset includes 199 dialogue transcriptions and videos captured in a multi-player social deduction game setting and 26,647 utterance level annotations of persuasion strategy and game level annotation of deduction game outcomes. |
OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens (2025.acl-demo)
Copied to clipboard
Jiacheng Liu, Taylor Blanton, Yanai Elazar, Sewon Min, Yen-Sung Chen, Arnavi Chheda-Kothary, Huy Tran, Byron Bischoff, Eric Marsh, Michael Schmitz, Cassidy Trier, Aaron Sarnat, Jenna James, Jon Borchardt, Bailey Kuehl, Evie Yu-Yen Cheng, Karen Farley, Taira Anderson, David Albright, Carissa Schoenick, Luca Soldaini, Dirk Groeneveld, Rock Yuren Pang, Pang Wei Koh, Noah A. Smith, Sophie Lebrecht, Yejin Choi, Hannaneh Hajishirzi, Ali Farhadi, Jesse Dodge
| Challenge: | tracing language models' outputs back to training data is a problem because they are trained on text corpora with trillions of tokens . existing methods for tracers have not been scaled to work within this multi-trillion-token setting . |
| Approach: | They propose a system that traces language models' outputs verbatim back to training data . OLMOTRACE retrieves documents from the model's training data that contain exact matches . |
| Outcome: | The proposed system can find verbatim matches between LM output and training data . it can be used to explore fact checking, hallucination, and creativity of language models . |
Improving Zero-Shot Translation by Disentangling Positional Information (2021.acl-long)
Copied to clipboard
| Challenge: | Multilingual neural machine translation has shown the capability of directly translating between language pairs unseen in training, i.e. zero-shot translation. |
| Approach: | They propose to remove residual connections in an encoder layer to reduce the difficulty of generalizing to new translation directions. |
| Outcome: | The proposed model outperforms pivot-based translation in terms of quality and ease of integration of new languages. |
Taylor Unswift: Secured Weight Release for Large Language Models via Taylor Expansion (2024.emnlp-main)
Copied to clipboard
Guanchu Wang, Yu-Neng Chuang, Ruixiang Tang, Shaochen Zhong, Jiayi Yuan, Hongye Jin, Zirui Liu, Vipin Chaudhary, Shuai Xu, James Caverlee, Xia Hu
| Challenge: | Existing mechanisms compromise ownership rights or raise data privacy concerns . existing mechanisms compromise security of released large language models . |
| Approach: | They propose a TaylorMLP to preserve the ownership of large language models by transforming the weights of LLMs into Taylor-series parameters instead of releasing original weights . |
| Outcome: | The proposed model preserves ownership of large language models and prevents their abuse by adjusting the generation speed and causing low-speed token generation. |
DisastQA: A Comprehensive Benchmark for Evaluating Question Answering in Disaster Management (2026.findings-acl)
Copied to clipboard
Zhitong Chen, Kai Yin, Xiangjue Dong, Chengkai Liu, Xiangpeng Li, Bo Li, Junwei Ma, Yiming Xiao, Ali Mostafavi, James Caverlee
| Challenge: | Existing benchmarks for question answering (QA) are lacking in a high-stakes environment. |
| Approach: | They propose a rigorously verified benchmark of 3,000 expert-annotated questions . they propose 'keypoint-based evaluation protocol' emphasizing factual completeness over verbosity . |
| Outcome: | Experiments with 20 models reveal substantial divergences from general-purpose models such as MMLU-Pro. |
Controllable and Reliable Knowledge-Intensive Task-Oriented Conversational Agents with Declarative Genie Worksheets (2025.acl-long)
Copied to clipboard
| Challenge: | Existing LLMs suffer from hallucination, following instructions with conditional logic, and integrating knowledge from different sources. |
| Approach: | They propose a programmable framework for creating knowledge-intensive task-oriented conversational agents that handle involved interactions and answer complex queries. |
| Outcome: | The proposed framework outperforms SOTA methods on complex logic dialogue datasets by up to 20.5%. |
OctoTools: A Multi-Agent Framework with Extensible Tools for Complex Reasoning (2026.acl-long)
Copied to clipboard
| Challenge: | Existing prompting methods for large language models (LLMs) are restricted to specialized domains, limited tool types, or require additional training data. |
| Approach: | They propose a training-free, user-friendly, and easily extensible multi-agent framework designed to tackle complex reasoning across diverse domains. |
| Outcome: | The proposed framework outperforms AutoGen, GPT-Functions, and LangChain by up to 10.6% when given the same set of tools. |
Defending Against Social Engineering Attacks in the Age of LLMs (2024.emnlp-main)
Copied to clipboard
Lin Ai, Tharindu Kumarage, Amrita Bhattacharjee, Zizhou Liu, Zheng Hui, Michael Davinroy, James Cook, Laura Cassani, Kirill Trapeznikov, Matthias Kirchner, Arslan Basharat, Anthony Hoogs, Joshua Garland, Huan Liu, Julia Hirschberg
| Challenge: | Existing research has developed frameworks to understand human-to-human CSE attacks. |
| Approach: | They propose a modular defense pipeline that improves detection at both the message and conversation levels. |
| Outcome: | The proposed model can be exploited to facilitate chat-based social engineering attacks and generate high-quality CSE content, but their detection capabilities are suboptimal, leading to increased operational costs for defense. |
Corrupted but Not Broken: Understanding and Mitigating the Negative Impacts of Corrupted Data in Visual Instruction Tuning (2025.emnlp-main)
Copied to clipboard
Yunhao Gou, Hansi Yang, Zhili Liu, Kai Chen, Yihan Zeng, Lanqing Hong, Zhenguo Li, Qun Liu, Bo Han, James Kwok, Yu Zhang
| Challenge: | Visual Instruction Tuning (VIT) aims to enhance Multimodal Large Language Models (MLLMs), but its effectiveness is often compromised by corrupted datasets with issues such as hallucinated content and poor OCR quality. |
| Approach: | They propose a corruption-robust training paradigm that surpasses existing strategies for mitigating the effects of corrupted data. |
| Outcome: | The proposed training paradigm surpasses existing strategies for mitigating the effects of corrupted data. |
Cross-Modal Discrete Representation Learning (2022.acl-long)
Copied to clipboard
| Challenge: | a new framework for learning representations from multimodal data is proposed . the proposed framework uses discretized embedding vectors to capture finer levels of granularity . |
| Approach: | They propose a self-supervised representation learning framework that captures finer levels of granularity across different modalities. |
| Outcome: | The proposed representation can capture finer levels of granularity across different modalities . it can be used on cross-modal retrieval tasks without direct supervision . |
GraphCheck: Breaking Long-Term Text Barriers with Extracted Knowledge Graph-Powered Fact-Checking (2025.acl-long)
Copied to clipboard
Yingjian Chen, Haoran Liu, Yinhong Liu, Jinxiang Xie, Rui Yang, Han Yuan, Yanran Fu, Peng Yuan Zhou, Qingyu Chen, James Caverlee, Irene Li
| Challenge: | Existing fact-checking methods that use large language models often generate subtle factual errors. |
| Approach: | They propose a fact-checking framework that uses extracted knowledge graphs to enhance text representation. |
| Outcome: | GraphCheck outperforms existing specialized fact-checkers on seven benchmarks spanning general and medical domains . Graph Neural Networks process extracted knowledge graphs as a soft prompt, enabling efficient fact- checking in a single inference call. |
Learning to Extract Structured Entities Using Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | Language Models (LMs) play a pivotal role in extracting structured information from unstructured text. |
| Approach: | They propose to reformulate the task to be entity-centric, enabling the use of diverse metrics that can provide more insights from various perspectives. |
| Outcome: | The proposed model outperforms baselines and human evaluations on the extracted entities. |
Eeyore: Realistic Depression Simulation via Expert-in-the-Loop Supervised and Preference Optimization (2025.findings-acl)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been explored for mental healthcare training and therapy client simulation, but they fail to authentically capture diverse client traits and psychological conditions. |
| Approach: | They propose an 8B model optimized for realistic depression simulation with expert input at every stage. |
| Outcome: | The model outperforms GPT-4o in linguistic authenticity and profile adherence. |
M2PT: Multimodal Prompt Tuning for Zero-shot Instruction Learning (2024.emnlp-main)
Copied to clipboard
Taowen Wang, Yiyang Liu, James Liang, Junhan Zhao, Yiming Cui, Yuning Mao, Shaoliang Nie, Jiahao Liu, Fuli Feng, Zenglin Xu, Cheng Han, Lifu Huang, Qifan Wang, Dongfang Liu
| Challenge: | Multimodal Large Language Models (MLLMs) exhibit remarkable performance across a wide range of domains. |
| Approach: | They propose a multimodal prompt tuning approach for efficient instruction tuning of MLLMs. |
| Outcome: | The proposed approach shows superior performance on multimodal evaluation datasets compared to state-of-the-art methods. |
COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation (2021.naacl-demos)
Copied to clipboard
Qingyun Wang, Manling Li, Xuan Wang, Nikolaus Parulian, Guangxing Han, Jiawei Ma, Jingxuan Tu, Ying Lin, Ranran Haoran Zhang, Weili Liu, Aabhas Chauhan, Yingjun Guan, Bangzheng Li, Ruisong Li, Xiangchen Song, Yi Fung, Heng Ji, Jiawei Han, Shih-Fu Chang, James Pustejovsky, Jasmine Rah, David Liem, Ahmed ELsayed, Martha Palmer, Clare Voss, Cynthia Schneider, Boyan Onyshkevych
| Challenge: | a new framework to digest relevant biomedical knowledge is needed to combat COVID-19 . quantity of research results is a bottleneck, and false information promoted in publications . |
| Approach: | a team of researchers has developed a framework to extract multimedia knowledge elements from scientific literature to combat COVID-19. |
| Outcome: | a new framework extracts fine-grained multimedia knowledge elements from scientific literature . it provides detailed contextual sentences, subfigures, and knowledge subgraphs as evidence . the framework is based on a case study of drug repurposing . |
DMRetriever: A Family of Models for Improved Text Retrieval in Disaster Management (2026.acl-long)
Copied to clipboard
| Challenge: | Existing models fail to handle the varied search intents inherent to disaster management scenarios, resulting in inconsistent and unreliable performance. |
| Approach: | They propose a new series of dense retrieval models tailored for disaster management that train on a three-stage framework with unsupervised contrastive pre-training and difficulty-aware progressive instruction fine-tuning. |
| Outcome: | The proposed model outperforms baseline models by 13.3 times and 33 times over baselines with only 7.6% of their parameters. |
Analyzing the Forgetting Problem in Pretrain-Finetuning of Open-domain Dialogue Response Models (2021.eacl-main)
Copied to clipboard
| Challenge: | a large-scale unsupervised pretraining has been shown to greatly boost the performance of natural language processing models. |
| Approach: | They propose an intuitive finetuning strategy to regularize the finetune process . they propose a mix-review strategy to alleviate the forgetting problem . |
| Outcome: | The proposed strategy regularizes the finetuning process, and the forgetting problem is alleviated . the proposed strategy also improves the performance of the resulting model . |
More Samples or More Prompts? Exploring Effective Few-Shot In-Context Learning for LLMs with In-Context Sampling (2024.findings-naacl)
Copied to clipboard
Bingsheng Yao, Guiming Chen, Ruishi Zou, Yuxuan Lu, Jiachen Li, Shao Zhang, Yisi Sang, Sijia Liu, James Hendler, Dakuo Wang
| Challenge: | Existing studies on LLM prompting focus on selecting a better set of data samples inside one single prompt input, but why not design and leverage multiple ICL prompts together to further improve the LLM’s performance? |
| Approach: | They propose a low-resource LLM prompting technique to optimize the construction of multiple ICL prompt inputs to produce confident predictions. |
| Outcome: | The proposed technique can produce confident predictions by optimizing the construction of multiple ICL prompt inputs on four NLI datasets and one QA dataset. |
Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework (2026.findings-acl)
Copied to clipboard
| Challenge: | Visual Document Retrieval (VDR) is of importance in multimodal retrieval applications. |
| Approach: | They propose a two-stage pruning and merging frameworks that combine pruning and merge techniques to achieve higher compression rates. |
| Outcome: | The proposed framework outperforms existing methods on 29 visual document retrieval datasets. |
Development and Benchmarking of a Blended Human-AI Qualitative Research Assistant (2026.acl-industry)
Copied to clipboard
Joseph Matveyenko, James Liu, John David Parsons, Ryan Brown, Alina I. Palimaru, Vipul Gupta, Prateek Puri
| Challenge: | Qualitative research emphasizes constructing meaning through iterative engagement with textual data. |
| Approach: | They present and benchmark a qualitative research assistant system that allows researchers to identify themes and annotate datasets. |
| Outcome: | The proposed system achieves an inter-rater reliability between Muse and humans of Cohen’s = 0.7 for well-specified codes. |
Detecting LLM-Assisted Cheating on Open-Ended Writing Tasks on Language Proficiency Tests (2024.emnlp-industry)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been used for open-ended writing tasks . however, there are limitations in detecting LLM-generated samples . |
| Approach: | They propose a framework for training LLM-generated text detectors that can detect LLM generated samples after being copy-typed. |
| Outcome: | The proposed model outperforms the transformer-based classifier on a high-stakes online English proficiency test. |
Mixture of insighTful Experts (MoTE): The Synergy of Reasoning Chains and Expert Mixtures in Self-Alignment (2025.acl-long)
Copied to clipboard
Zhili Liu, Yunhao Gou, Kai Chen, Lanqing Hong, Jiahui Gao, Fei Mi, Yu Zhang, Zhenguo Li, Xin Jiang, Qun Liu, James Kwok
| Challenge: | Recent studies show that reasoning abilities contribute significantly to model safety, while integrating Mixture-of-Experts (MoE) architectures can further enhance alignment. |
| Approach: | They propose a framework that synergistically combines reasoning chains and expert mixtures to improve self-alignment. |
| Outcome: | The proposed framework improves model safety, jailbreak resistance, and over-refusal capabilities, achieving performance comparable to OpenAI’s state-of-the-art o1 model. |
Forward-Backward Reasoning in Large Language Models for Mathematical Verification (2024.findings-acl)
Copied to clipboard
| Challenge: | Extensive experiments on six standard mathematical data sets and three LLMs show that FOBAR achieves state-of-the-art performance. |
| Approach: | They propose to combine forward and backward reasoning to verify candidate answers . they propose to use a template to mask a number and ask the LLM to answer a backward question . |
| Outcome: | Experiments on mathematical data show that proposed backward reasoning outperforms Self-Consistency. |