Added author page for Yuqing Yang

data/xml/2021.emnlp.xml

@@ -2827,7 +2827,7 @@
       <title>Uncertain Local-to-Global Networks for Document-Level Event Factuality Identification</title>
       <author><first>Pengfei</first><last>Cao</last></author>
       <author><first>Yubo</first><last>Chen</last></author>
-      <author><first>Yuqing</first><last>Yang</last></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last></author>
       <author><first>Kang</first><last>Liu</last></author>
       <author><first>Jun</first><last>Zhao</last></author>
       <pages>2636–2645</pages>

data/xml/2022.findings.xml

@@ -11467,7 +11467,7 @@ Faster and Smaller Speech Translation without Quality Compromise</title>
     <paper id="253">
       <title><fixed-case>DORE</fixed-case>: Document Ordered Relation Extraction based on Generative Framework</title>
       <author><first>Qipeng</first><last>Guo</last><affiliation>Amazon Shanghai AI Lab</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
       <author><first>Hang</first><last>Yan</last><affiliation>Fudan University</affiliation></author>
       <author><first>Xipeng</first><last>Qiu</last><affiliation>Fudan University</affiliation></author>
       <author><first>Zheng</first><last>Zhang</last><affiliation>NYU Shanghai</affiliation></author>

data/xml/2023.acl.xml

@@ -10376,7 +10376,7 @@
     </paper>
     <paper id="720">
       <title>An <fixed-case>AMR</fixed-case>-based Link Prediction Approach for Document-level Event Argument Extraction</title>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
       <author><first>Qipeng</first><last>Guo</last><affiliation>Amazon Shanghai AI Lab</affiliation></author>
       <author><first>Xiangkun</first><last>Hu</last><affiliation>Amazon</affiliation></author>
       <author><first>Yue</first><last>Zhang</last><affiliation>Westlake University</affiliation></author>

data/xml/2023.emnlp.xml

@@ -2379,7 +2379,7 @@
       <title>Plan, Verify and Switch: Integrated Reasoning with Diverse <fixed-case>X</fixed-case>-of-Thoughts</title>
       <author><first>Tengxiao</first><last>Liu</last></author>
       <author><first>Qipeng</first><last>Guo</last></author>
-      <author><first>Yuqing</first><last>Yang</last></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last></author>
       <author><first>Xiangkun</first><last>Hu</last></author>
       <author><first>Yue</first><last>Zhang</last></author>
       <author><first>Xipeng</first><last>Qiu</last></author>
@@ -11496,7 +11496,7 @@
       <author><first>Huiqiang</first><last>Jiang</last></author>
       <author><first>Qianhui</first><last>Wu</last></author>
       <author><first>Chin-Yew</first><last>Lin</last></author>
-      <author><first>Yuqing</first><last>Yang</last></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last></author>
       <author><first>Lili</first><last>Qiu</last></author>
       <pages>13358-13376</pages>
       <abstract>Large language models (LLMs) have been applied in various applications due to their astonishing capabilities. With advancements in technologies such as chain-of-thought (CoT) prompting and in-context learning (ICL), the prompts fed to LLMs are becoming increasingly lengthy, even exceeding tens of thousands of tokens. To accelerate model inference and reduce cost, this paper presents LLMLingua, a coarse-to-fine prompt compression method that involves a budget controller to maintain semantic integrity under high compression ratios, a token-level iterative compression algorithm to better model the interdependence between compressed contents, and an instruction tuning based method for distribution alignment between language models. We conduct experiments and analysis over four datasets from different scenarios, i.e., GSM8K, BBH, ShareGPT, and Arxiv-March23; showing that the proposed approach yields state-of-the-art performance and allows for up to 20x compression with little performance loss.</abstract>
@@ -15516,7 +15516,7 @@ The experiments were repeated and the tables and figures were updated. Changes a
       <author><first>Jiawei</first><last>Hong</last><affiliation>Fudan University</affiliation></author>
       <author><first>Keyu</first><last>Chen</last><affiliation>Fudan University</affiliation></author>
       <author><first>Xiaoran</first><last>Liu</last><affiliation>Fudan University</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last><affiliation>Fudan University</affiliation></author>
       <author><first>Honglin</first><last>Guo</last><affiliation>Fudan University</affiliation></author>
       <author><first>Tengxiao</first><last>Liu</last><affiliation>Fudan University</affiliation></author>
       <author><first>Yu</first><last>Sun</last><affiliation>Fudan University</affiliation></author>

data/xml/2024.acl.xml

@@ -1287,7 +1287,7 @@
       <author><first>Xufang</first><last>Luo</last><affiliation>Microsoft Research</affiliation></author>
       <author id="dongsheng-li-fudan"><first>Dongsheng</first><last>Li</last></author>
       <author><first>Chin-Yew</first><last>Lin</last><affiliation>Microsoft</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft</affiliation></author>
       <pages>1658-1677</pages>
       <abstract>In long context scenarios, large language models (LLMs) face three main challenges: higher computational cost, performance reduction, and position bias. Research indicates that LLM performance hinges on the density and position of key information in the input prompt. Inspired by these findings, we propose LongLLMLingua for prompt compression towards improving LLMs’ perception of the key information to simultaneously address the three challenges. Our extensive evaluation across various long context scenarios demonstrates that LongLLMLingua not only enhances performance but also significantly reduces costs and latency. For instance, in the NaturalQuestions benchmark, LongLLMLingua boosts performance by up to 21.4% with around 4x fewer tokens in GPT-3.5-Turbo, leading to substantial cost savings. It achieves a 94.0% cost reduction in the LooGLE benchmark. Moreover, when compressing prompts of about 10k tokens at ratios of 2x-6x, LongLLMLingua can accelerate end-to-end latency by 1.4x-2.6x.</abstract>
@@ -4308,7 +4308,7 @@
       <author><first>Qiyang</first><last>Jiang</last><affiliation>ShanghaiTech University</affiliation></author>
       <author><first>XingyuHan</first><last>XingyuHan</last></author>
       <author><first>Nan</first><last>Chen</last><affiliation>Microsoft</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author><first>Kan</first><last>Ren</last><affiliation>ShanghaiTech University</affiliation></author>
       <pages>5677-5700</pages>
       <abstract>In the era of data-driven decision-making, the complexity of data analysis necessitates advanced expertise and tools of data science, presenting significant challenges even for specialists. Large Language Models (LLMs) have emerged as promising aids as data science agents, assisting humans in data analysis and processing. Yet their practical efficacy remains constrained by the varied demands of real-world applications and complicated analytical process. In this paper, we introduce DSEval – a novel evaluation paradigm, as well as a series of innovative benchmarks tailored for assessing the performance of these agents throughout the entire data science lifecycle. Incorporating a novel bootstrapped annotation method, we streamline dataset preparation, improve the evaluation coverage, and expand benchmarking comprehensiveness. Our findings uncover prevalent obstacles and provide critical insights to inform future advancements in the field.</abstract>
@@ -6158,7 +6158,7 @@
     <paper id="445">
       <title>Full Parameter Fine-tuning for Large Language Models with Limited Resources</title>
       <author><first>Kai</first><last>Lv</last></author>
-      <author><first>Yuqing</first><last>Yang</last></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last></author>
       <author><first>Tengxiao</first><last>Liu</last><affiliation>Fudan University and Amazon</affiliation></author>
       <author><first>Qipeng</first><last>Guo</last><affiliation>Shanghai AI Laboratory</affiliation></author>
       <author><first>Xipeng</first><last>Qiu</last><affiliation>Fudan University</affiliation></author>

data/xml/2024.eacl.xml

@@ -2455,7 +2455,7 @@
       <author><first>Yuge</first><last>Zhang</last><affiliation>Microsoft</affiliation></author>
       <author><first>Kan</first><last>Ren</last><affiliation>Microsoft</affiliation></author>
       <author id="dongsheng-li-fudan"><first>Dongsheng</first><last>Li</last></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <pages>2931-2959</pages>
       <abstract>The field of machine learning (ML) has gained widespread adoption, leading to significant demand for adapting ML to specific scenarios, which is yet expensive and non-trivial. The predominant approaches towards the automation of solving ML tasks (e.g., AutoML) are often time-consuming and hard to understand for human developers. In contrast, though human engineers have the incredible ability to understand tasks and reason about solutions, their experience and knowledge are often sparse and difficult to utilize by quantitative approaches. In this paper, we aim to bridge the gap between machine intelligence and human knowledge by introducing a novel framework MLCopilot, which leverages the state-of-the-art large language models to develop ML solutions for novel tasks. We showcase the possibility of extending the capability of LLMs to comprehend structured inputs and perform thorough reasoning for solving novel ML tasks. And we find that, after some dedicated design, the LLM can (i) observe from the existing experiences of ML tasks and (ii) reason effectively to deliver promising results for new tasks. The solution generated can be used directly to achieve high levels of competitiveness.</abstract>
       <url hash="24fa996c">2024.eacl-long.179</url>

data/xml/2024.emnlp.xml

@@ -5875,7 +5875,7 @@
       <author><first>Zhiyuan</first><last>He</last><affiliation>Microsoft</affiliation></author>
       <author orcid="0000-0002-1327-4882"><first>Huiqiang</first><last>Jiang</last><affiliation>Microsoft</affiliation></author>
       <author><first>Zilong</first><last>Wang</last><affiliation>Microsoft Research</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author><first>Luna K.</first><last>Qiu</last><affiliation>Microsoft</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft</affiliation></author>
       <pages>7333-7345</pages>

data/xml/2024.findings.xml

@@ -6904,7 +6904,7 @@
       <author><first>Jue</first><last>Zhang</last><affiliation>Microsoft</affiliation></author>
       <author orcid="0000-0003-2559-2383"><first>Qingwei</first><last>Lin</last><affiliation>Microsoft Research</affiliation></author>
       <author orcid="0000-0002-8957-7628"><first>Victor</first><last>Rühle</last><affiliation>Microsoft</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author><first>Chin-Yew</first><last>Lin</last><affiliation>Microsoft</affiliation></author>
       <author><first>H. Vicky</first><last>Zhao</last><affiliation>Tsinghua University, Tsinghua University</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft</affiliation></author>
@@ -26114,7 +26114,7 @@ and high variation in performance on the subset, suggesting our plausibility cri
     </paper>
     <paper id="490">
       <title>Weak-to-Strong Reasoning</title>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>University of Southern California</affiliation></author>
+      <author id="yuqing-yang-usc"><first>Yuqing</first><last>Yang</last><affiliation>University of Southern California</affiliation></author>
       <author orcid="0009-0000-9262-3771"><first>Yan</first><last>Ma</last></author>
       <author><first>Pengfei</first><last>Liu</last></author>
       <pages>8350-8367</pages>
@@ -29668,7 +29668,7 @@ hai-coaching/</abstract>
       <author><first>Fangyun</first><last>Wei</last></author>
       <author><first>Zongqing</first><last>Lu</last><affiliation>Tsinghua University, Tsinghua University</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <pages>12806-12816</pages>
       <abstract>Efficient fine-tuning plays a fundamental role in modern large models, with low-rank adaptation emerging as a particularly promising approach. However, the existing variants of LoRA are hampered by limited expressiveness, a tendency to overfit, and sensitivity to hyperparameter settings. This paper presents LoRA Slow Cascade Learning (LoRASC), an innovative technique designed to enhance LoRA’s expressiveness and generalization capabilities while preserving its training efficiency. Our approach augments expressiveness through a cascaded learning strategy that enables a mixture-of-low-rank adaptation, thereby increasing the model’s ability to capture complex patterns. Additionally, we introduce a slow-fast update mechanism and cascading noisy tuning to bolster generalization. The extensive experiments on various language and vision datasets, as well as robustness benchmarks, demonstrate that the proposed method not only significantly outperforms existing baselines, but also mitigates overfitting, enhances model stability, and improves OOD robustness.</abstract>
       <url hash="f9c1e27c">2024.findings-emnlp.748</url>

data/xml/2025.emnlp.xml

@@ -23272,7 +23272,7 @@
       <author><first>Zhiyuan</first><last>He</last><affiliation>Microsoft</affiliation></author>
       <author orcid="0000-0002-1327-4882"><first>Huiqiang</first><last>Jiang</last><affiliation>Microsoft</affiliation></author>
       <author><first>Chengruidong</first><last>Zhang</last><affiliation>Microsoft</affiliation></author>
-      <author orcid="0000-0003-3518-5212"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author orcid="0000-0003-3518-5212" id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author orcid="0000-0002-7555-170X"><first>Jianyong</first><last>Wang</last><affiliation>Tsinghua University, Tsinghua University</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft and University of Texas at Austin</affiliation></author>
       <pages>31110-31125</pages>

data/xml/2025.findings.xml

@@ -10848,7 +10848,7 @@
       <author><first>Qianhui</first><last>Wu</last><affiliation>Microsoft</affiliation></author>
       <author><first>Chin-Yew</first><last>Lin</last><affiliation>Microsoft</affiliation></author>
       <author id="dongsheng-li-fudan" orcid="0000-0003-3103-8442"><first>Dongsheng</first><last>Li</last><affiliation>Microsoft Research Asia</affiliation></author>
-      <author orcid="0000-0003-3518-5212"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
+      <author orcid="0000-0003-3518-5212" id="yuqing-yang"><first>Yuqing</first><last>Yang</last><affiliation>Research, Microsoft</affiliation></author>
       <author orcid="0000-0003-3825-2230"><first>Yongfeng</first><last>Huang</last><affiliation>Tsinghua University, Tsinghua University</affiliation></author>
       <author><first>Lili</first><last>Qiu</last><affiliation>Microsoft and University of Texas at Austin</affiliation></author>
       <pages>6092-6111</pages>
@@ -29842,7 +29842,7 @@
       <author><first>Jiacheng</first><last>Li</last></author>
       <author><first/><last>Zoupanxiang</last></author>
       <author><first>Yudong</first><last>Zhou</last><affiliation>PriorShape and Richinfo</affiliation></author>
-      <author><first>Yuqing</first><last>Yang</last></author>
+      <author id="yuqing-yang"><first>Yuqing</first><last>Yang</last></author>
       <pages>5069-5081</pages>
       <abstract>This study investigates the position bias in information retrieval, where models tend to overemphasize content at the beginning of passages while neglecting semantically relevant information that appears later. To analyze the extent and impact of position bias, we introduce a new evaluation framework consisting of two position-aware retrieval benchmarks (SQuAD-PosQ, FineWeb-PosQ) and an intuitive diagnostic metric, the Position Sensitivity Index (PSI), for quantifying position bias from a worst-case perspective. We conduct a comprehensive evaluation across the full retrieval pipeline, including BM25, dense embedding models, ColBERT-style late-interaction models, and full-interaction reranker models. Our experiments show that when relevant information appears later in the passage, dense embedding models and ColBERT-style models suffer significant performance degradation (an average drop of 15.6%). In contrast, BM25 and reranker models demonstrate greater robustness to such positional variation. These findings provide practical insights into model sensitivity to the position of relevant information and offer guidance for building more position-robust retrieval systems. Code and data are publicly available at: https://github.com/NovaSearch-Team/position-bias-in-IR.</abstract>
       <url hash="28cb088e">2025.findings-emnlp.271</url>

data/yaml/name_variants.yaml

@@ -605,6 +605,14 @@
 - canonical: {first: François, last: Barthélemy}
   variants:
   - {first: Francois, last: Barthelemy}
+- canonical: {first: Yuqing, last: Yang}
+  id: yuqing-yang-usc
+  orcid: 0009-0006-7205-1191
+  institution: University of Southern California
+  comment: USC
+- canonical: {first: Yuqing, last: Yang}
+  id: yuqing-yang
+  comment: May refer to several people
 - canonical: {first: G. Edward, last: Barton}
   variants:
   - {first: G. Edward, last: 'Barton, Jr.'}