Papers by Xiaoyuan Yi
MotiveBench: How Far Are We From Human-Like Motivational Reasoning in Large Language Models? (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks for large language models lack information asymmetry with real-world situations. |
| Approach: | They propose a benchmark to evaluate the human-like motivational and behavioral reasoning ability of LLMs with detailed, realistic situations. |
| Outcome: | The proposed benchmark compared LLMs with real-world scenarios on seven model families and found that the most advanced models struggle with understanding "love & belonging" needs. |
Can Persona-Prompted LLMs Emulate Subgroup Values? An Empirical Analysis of Generalisability and Fairness in Cultural Alignment (2026.acl-long)
Copied to clipboard
Bryan Chen Zhengyu Tan, Zhengyuan Liu, Xiaoyuan Yi, Jing Yao, Xing Xie, Nancy F. Chen, Roy Ka-Wei Lee
| Challenge: | Current alignment paradigms treat "human values" as a monolithic entity, ignoring the fact that many societies are a mosaic of diverse subgroups with distinct and sometimes conflicting values, preferences, and norms. |
| Approach: | They examine whether Large Language Models can emulate distinct cultural values of subgroups . they use a global value survey to examine the value landscape of a multicultural society . |
| Outcome: | The proposed model improves on unseen, out-of-distribution subgroups by 17.4% . the model widens the disparity between subgroup groups when measured by distance-aware metrics. |
Automatic Poetry Generation with Mutual Reinforcement Learning (D18-1)
Copied to clipboard
| Challenge: | Existing models for automatic poetry generation are based on maximum likelihood estimation (MLE) MLE-based models tend to remember common patterns of the poetry corpus, which results in loss-evaluation mismatch. |
| Approach: | They propose to model the criteria and use them as explicit rewards to guide gradient update by reinforcement learning to motivate the model to pursue higher scores. |
| Outcome: | The proposed model outperforms the current state-of-the-art model and improves on Chinese poetry. |
Unintended Harms of Value-Aligned LLMs: Psychological and Empirical Insights (2025.acl-long)
Copied to clipboard
| Challenge: | Value-aligned LLMs are more prone to harmful behavior than fine-tuned models . value-aligned models generate text according to the aligned values, which can amplify harmful outcomes. |
| Approach: | They propose to use in-context alignment methods to enhance the safety of value-aligned LLMs. |
| Outcome: | The proposed methods improve value alignment and safety, the authors say . value-aligned models are more prone to harmful behavior than fine-tuned models . |
MMAC: A Multilingual, Multimodal Alignment Framework for Cultural Grounding Evaluation (2026.acl-long)
Copied to clipboard
Weihua Zheng, Zhengyuan Liu, Tanmoy Chakraborty, Weiwen Xu, Xiaoxue Gao, Bryan Chen Zhengyu Tan, Bowei Zou, Chang Liu, Yujia Hu, Xing Xie, Xiaoyuan Yi, Jing Yao, Chaojun Wang, Long Li, Rui Liu, Huiyao Liu, Koji Inoue, Ryuichi Sumida, Tatsuya Kawahara, Fan Xu, Lingyu Ye, Wei Tian, Dongjun Kim, Jimin Jung, Jaehyung Seo, Nadya Yuki Wangsajaya, Pham Minh Duc, Ojasva Saxena, Palash Nandi, Xiyan Tao, Wiwik Karlina, Tuan Luong, Keertana Arun Vasan, Roy Ka-Wei Lee, Nancy F. Chen
| Challenge: | Existing models lack cultural alignment across modalities and languages . a new framework to assess cultural awareness across linguistics and languages is needed . |
| Approach: | They propose a framework that integrates tri-modally aligned cultural benchmarks and a five-dimensional evaluation protocol to assess cross-country awareness disparities. |
| Outcome: | The proposed framework assesses cultural awareness disparities across modalities and languages . it is the first dataset aligned at the input level across text, image, and speech . |
Evade the Trap of Mediocrity: Promoting Diversity and Novelty in Text Generation via Concentrating Attention (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies have shown that powerful Transformer architectures produce dull high-frequency phrases, severely hurting the diversity and novelty of generated text. |
| Approach: | They propose a method to control the sharpness of the attention distribution by python code and use it to learn a Bayesian approximation of posterior attention. |
| Outcome: | The proposed method improves diversity and novelty while maintaining comparable quality on conditional and unconditional generation tasks. |
Value Compass Benchmarks: A Comprehensive, Generative and Self-Evolving Platform for LLMs’ Value Evaluation (2025.acl-demo)
Copied to clipboard
Jing Yao, Xiaoyuan Yi, Shitong Duan, Jindong Wang, Yuzhuo Bai, Muhua Huang, Yang Ou, Scarlett Li, Peng Zhang, Tun Lu, Zhicheng Dou, Maosong Sun, James Evans, Xing Xie
| Challenge: | Current evaluation methods for large language models face two key challenges: 1. evaluation validity and 2. Result interpretation reduce the pluralistic and incommensurable values to one-dimensional scores. |
| Approach: | They propose a platform for comprehensive value diagnosis of large language models (LLMs) that provides a generative evaluation paradigm that automatically creates real-world test items co-evolving with ever-advancing LLMs. |
| Outcome: | The proposed platform provides a framework for comprehensive value diagnosis of large language models (LLMs) with fine-grained scores and case studies across 27 value dimensions for 33 leading LLMs, customized comparisons, and visualized analysis of LLM’s alignment with cultural values. |
DuNST: Dual Noisy Self Training for Semi-Supervised Controllable Text Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to augment self-training (ST) in attribute-controllable language generation are limited and limited. |
| Approach: | They propose a new ST framework that integrates self-generated pseudo text into attribute-controllable language generation. |
| Outcome: | The proposed framework can be applied to semi-supervised controllable language generation. |
Value FULCRA: Mapping Large Language Models to the Multidimensional Spectrum of Basic Human Value (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing work specifies values as risk criteria formulated in the AI community, e.g., fairness and privacy protection, suffering from poor clarity, adaptability and transparency. |
| Approach: | They propose a value alignment paradigm based on Schwartz's Theory of Basic Values as an instantiation and propose 'BaseAlign' to support this paradigm. |
| Outcome: | The proposed model covers existing risks and anticipates unidentified ones with a low-data set. |
ToViLaG: Your Visual-Language Generative Model is Also An Evildoer (2023.emnlp-main)
Copied to clipboard
| Challenge: | Recent large-scale Visual-Language Generative Models (VLGMs) generate toxic content, e.g., offensive text and pornography images, raising significant ethical risks. |
| Approach: | They propose a bottleneck-based detoxification method to reduce toxicity while maintaining comparable generation quality. |
| Outcome: | The proposed method could reduce toxicity while maintaining comparable generation quality. |
MoVa: Towards Generalizable Classification of Human Morals and Values (2025.emnlp-main)
Copied to clipboard
Ziyu Chen, Junfei Sun, Chenxi Li, Tuan Dung Nguyen, Jing Yao, Xiaoyuan Yi, Xing Xie, Chenhao Tan, Lexing Xie
| Challenge: | Identifying human morals and values embedded in language is essential to empirical studies of communication. |
| Approach: | They propose a framework for generalizable classification of human morals and values . they recommend a classification strategy that scores all related concepts simultaneously . |
| Outcome: | The proposed method outperforms fine-tuned models across domains and frameworks. |
Chunks as Arms: Multi-Armed Bandit-Guided Sampling for Long-Context LLM Preference Optimization (2026.acl-long)
Copied to clipboard
Shaohua Duan, Pengcheng Huang, Xinze Li, Zhenghao Liu, Xiaoyuan Yi, Yukun Yan, Shuo Wang, Yu Gu, Ge Yu, Maosong Sun
| Challenge: | Recent studies have explored fine-tuning Large Language Models with synthetic data to enhance their long-context capabilities. |
| Approach: | They propose a framework that leverages a Multi-Armed Bandit rollout strategy to identify the most informative chunks from the given long context for sampling high-quality and diverse responses. |
| Outcome: | The proposed framework achieves 4% improvement on long-context reasoning benchmarks on Llama and Qwen. |
Negating Negatives: Alignment with Human Negative Samples via Distributional Dispreference Optimization (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to steer LLMs towards human preference suffer from noisy positive-negative training pairs. |
| Approach: | They propose a distributional preference optimization method which maximizes discrepancy between dispreferred responses and generated non-negative ones. |
| Outcome: | The proposed method achieves comparable generation quality and surpasses the latest strong baselines in producing less harmful and more informative responses with better training stability and faster convergence. |
Recurrence Boosts Diversity! Revisiting Recurrent Latent Variable in Transformer-Based Variational AutoEncoder for Diverse Text Generation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Variational Auto-Encoder (VAE) has been widely adopted in text generation due to its ability to learn flexible representations. |
| Approach: | They propose a Transformer-based recurrent VAE structure that imposes recurrence on segment-wise latent variables with arbitrarily separated text segments and constructs the posterior distribution with residual parameterization. |
| Outcome: | The proposed structure can deduce a non-zero lower bound of the KL term and enhance the entanglement of each segment and preceding latent variables, providing a theoretical guarantee of generation diversity. |
Influence-based Online Experience Selection for Effective RLHF (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for RL fail to establish an interpretable connection between data and optimization objectives. |
| Approach: | They propose a data selection method that dynamically estimates the influence of individual training samples on policy optimization. |
| Outcome: | The proposed method significantly improves training effectiveness with fewer optimization steps. |
Stylistic Chinese Poetry Generation via Unsupervised Style Disentanglement (D18-1)
Copied to clipboard
| Challenge: | Automatic Chinese poetry generation is one of the first attempts towards computer writing. |
| Approach: | They propose a model which requires no supervised style labeling to generate stylistic poems . they incorporate mutual information, a concept in information theory, into modeling . |
| Outcome: | The proposed model generates stylistic poems without losing fluency and coherency . it is based on mutual information, a concept in information theory . |
Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing GEC models produce spurious corrections or fail to detect lots of errors. |
| Approach: | They propose a neural network for GEC quality estimation with multiple hypotheses . VERNet establishes interactions among hypothese based on reasoning graph . |
| Outcome: | The proposed model achieves state-of-the-art grammatical error detection performance and best quality estimation results on four GEC datasets. |
Jiuge: A Human-Machine Collaborative Chinese Classical Poetry Generation System (P19-3)
Copied to clipboard
Guo Zhipeng, Xiaoyuan Yi, Maosong Sun, Wenhao Li, Cheng Yang, Jiannan Liang, Huimin Chen, Yuhui Zhang, Ruoyu Li
| Challenge: | Existing systems for automatic poetry generation are model-oriented, resulting in poor user participation. |
| Approach: | They propose a human-machine collaborative Chinese classical poetry generation system called Jiuge . Jiuge allows users to revise unsatisfied parts of a generated poem draft repeatedly . |
| Outcome: | The proposed system allows users to revise unsatisfied parts of a generated poem draft repeatedly. |
Fuse It More Deeply! A Variational Transformer with Layer-Wise Latent Variable Inference for Text Generation (2022.naacl-main)
Copied to clipboard
| Challenge: | Variational Auto-Encoders are often used for text generation tasks due to the sequential nature of the text. |
| Approach: | They propose a variational Transformer framework that learns a series of layer-wise latent variables with each inferred from those of lower layers and tightly coupled with the hidden states by low-rank tensor product. |
| Outcome: | The proposed framework can learn latent variables from lower layers and incorporate more information. |
Generative Personality Simulation via Theory-Informed Structured Interview (2026.eacl-long)
Copied to clipboard
Pengda Wang, Huiqi Zou, Han Jiang, Hanjie Chen, Tianjun Sun, Xiaoyuan Yi, Ziang Xiao, Frederick L. Oswald
| Challenge: | Personality structured interviews are often lacking in advancing social science research. |
| Approach: | They propose a method to incorporate psychological insights into LLM simulations . they use a measure theory grounded evaluation procedure to evaluate reliability and validity . |
| Outcome: | The proposed method improves human-like heterogeneity in LLM-simulated personality data and predicts personality-related behavioral outcomes. |
Towards Better Value Principles for Large Language Model Alignment: A Systematic Evaluation and Enhancement (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) show remarkable performance across tasks . alignment with human values is critical for their responsible development. |
| Approach: | They propose a framework that evaluates value principles along three desirable properties . they propose supervised fine-tuning, reinforcement learning-based approaches . |
| Outcome: | The proposed framework improves value principles along the three desirable properties of LLMs. |