Papers by Kaiyuan Liu
Equipping Retrieval-Augmented Large Language Models with Document Structure Awareness (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to retrieval-augmented generation ignore valuable structure that is crucial for document organization. |
| Approach: | They propose a framework that explicitly incorporates structural information throughout the RAG process. |
| Outcome: | The proposed framework incorporates structural information throughout the RAG process. |
MARS-Bench: A Multi-turn Athletic Real-world Scenario Benchmark for Dialogue Evaluation (2025.findings-emnlp)
Copied to clipboard
Chenghao Yang, Yinbo Luo, Zhoufutu Wen, Qi Chu, Tao Gong, Longxiang Liu, Kaiyuan Zhang, Jianpeng Jiao, Ge Zhang, Wenhao Huang, Nenghai Yu
| Challenge: | Large Language Models (LLMs) have been widely adopted in real-world dialogue applications, but their robustness is criticized all along. |
| Approach: | They propose to use play-by-play text commentary to build a multi-turn athletic real-world scenario dialogue benchmark to evaluate three critical aspects of multi-turned conversations: ultra multi- turn, interactive multi-twist, and cross-turn tasks. |
| Outcome: | The proposed benchmarks outperform open-source LLMs on three critical aspects of multi-turn conversations: ultra multi-turned, interactive multi- turn, and cross-turn tasks. |
MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus (2026.findings-acl)
Copied to clipboard
Yexing Du, Kaiyuan Liu, Bihe Zhang, Youcheng Pan, Bo Yang, Liangyu Huo, Xiyuan Zhang, Jian Xie, Daojing He, Yang Xiang, Ming Liu, Bing Qin
| Challenge: | Multimodal Large Language Models (MLLMs) have advanced Chinese Classical Studies (CCS) but the audio dimension of CCS remains underexplored due to a lack of high-quality, domain-specific audio corpora. |
| Approach: | They propose a 119-hour audio corpus comprising 22,000 audio samples to bridge this gap . it encompasses a diverse range of literary genres across six tasks . |
| Outcome: | The proposed corpus encompasses a diverse range of literary genres across six tasks: Automatic Speech Recognition (ASR), Speech-to-Text Translation (S2TT), Speech Emotion Captioning (SEC), Spoken Question Answering ( SQA), Speech Understanding (SU), and Speech Reasoning (SR). |
ProjectEval: A Benchmark for Programming Agents Automated Evaluation on Project-Level Code Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing benchmarks lack the ability to automatically evaluate from users’ perspective and lack the explainability of the results of LLM agents’ code generation capabilities. |
| Approach: | They propose a new benchmark for LLM agents' automated evaluation by simulating user interaction. |
| Outcome: | The proposed benchmark can evaluate the generated projects by user interaction simulation and by code similarity through existing objective indicators. |
On the Step Length Confounding in LLM Reasoning Data Selection (2026.findings-acl)
Copied to clipboard
Bing Wang, Rui Miao, Chen Shen, Shaotian Yan, Kaiyuan Liu, Ximing Li, Xiaosong Yuan, Sinan Fan, Jun Zhang, Jieping Ye
| Challenge: | Existing pipelines generate long reasoning data from more capable Large Language Models (LLMs) and apply manually heuristic or naturalness-based selection methods to filter high-quality samples. |
| Approach: | They propose to use supervised fine-tuning to generate long reasoning data from more capable Large Language Models and apply manually heuristic or naturalness-based selection methods to filter high-quality samples. |
| Outcome: | Experiments on four LLMs and five evaluation benchmarks show that the proposed approach is effective in mitigating step length confounding problem. |
LIST: Linearly Incremental SQL Translator for Single-Hop Reasoning, Generation and Verification (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing schema linking methods are not able to handle complex SQL queries. |
| Approach: | They propose a new algorithm that transforms SQL queries into grammatically verifiable sub-queries which are arranged sequentially to reflect single-hop reasoning steps. |
| Outcome: | The proposed algorithm achieves significant performance gains on the BIRD dataset and surpasses schema linking methods at comparable or better cost. |
GRASS: Gradient-based Adaptive Layer-wise Importance Sampling for Memory-efficient Large Language Model Fine-tuning (2026.findings-acl)
Copied to clipboard
Kaiyuan Tian, Linbo Qiao, Yu Tang, Gongqingjian Jiang, Baihui Liu, Yifu Gao, Xialin Su, Dongsheng Li
| Challenge: | Low-rank adaptation methods for large language models limit expressiveness and performance . layer-wise fine-tuning methods overlook variations in layer importance across tasks and training stages, resulting in suboptimal performance on downstream tasks. |
| Approach: | They propose a gradient-based adaptive layer-wise importance sampling framework that updates only a subset of parameters to reduce memory usage. |
| Outcome: | The proposed framework outperforms state-of-the-art methods in accuracy and memory usage. |
Alloc-MoE: Budget-Aware Expert Activation Allocation for Efficient Mixture-of-Experts Inference (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches that reduce expert activations lead to severe model performance degradation. |
| Approach: | They propose a framework that optimizes budget allocation coordinately at layer and token levels to minimize model performance degradation. |
| Outcome: | The proposed framework achieves 1.15 prefill and 1.34 decode speedups on DeepSeek-V2-Lite at half of the original budget. |
Take Its Essence, Discard Its Dross! Debiasing for Toxic Language Detection via Counterfactual Causal Effect (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods to mitigate lexical bias in toxic language detection (TLD) do not exploit the “useful” and “misleading” impact of the bias. |
| Approach: | They propose a counterfactual Causal Debiasing Framework to mitigate lexical bias in toxic language detection (TLD) it preserves the “useful impact” of lexical bias and eliminates the "misleading impact" they propose to use the same framework to analyze the causal effect of a sentence and bias tokens . |
| Outcome: | The proposed framework preserves the “useful impact” of lexical bias and eliminates the ‘misleading impact’ Empirical evaluations show that the proposed model outperforms current debiased models for out-of-distribution data. |
Chinese Court Simulation with LLM-Based Agents System (2026.findings-acl)
Copied to clipboard
Kaiyuan Zhang, Jiaqi Li, Yueyue Wu, Haitao Li, Cheng Luo, Shaokun Zou, Yujia Zhou, Weihang Su, Yiqun Liu, Qingyao Ai
| Challenge: | Existing studies have neglected the systematic design and procedure evaluation of court simulations, which are critical to the credibility and usage of court simulators in practice. |
| Approach: | They propose a court simulation paradigm based on the real-world procedure structure of Chinese courts and a framework that focuses on both legal judgment prediction and court procedure analysis. |
| Outcome: | The proposed model outperforms judges and lawyers from the real trials in many aspects. |
MADAWSD: Multi-Agent Debate Framework for Adversarial Word Sense Disambiguation (2025.emnlp-main)
Copied to clipboard
Kaiyuan Zhang, Qian Liu, Luyang Zhang, Chaoqun Zheng, Shuaimin Li, Bing Xu, Muyun Yang, Xinxiao Qiao, Wenpeng Lu
| Challenge: | Word sense disambiguation (WSD) is a fundamental yet challenging task in natural language processing. |
| Approach: | a novel multi-agent Debate framework for adversarial word Sense disambiguation is proposed . the framework simulates a real-world debate environment where multiple agents engage in discussions about ambiguous words in the context of adversarials. |
| Outcome: | The proposed framework integrates with existing LLMs and improves models in Chinese language . it shows that it can be used to improve models in the Chinese language and improve performance . |