Yuxuan Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Yuxuan Liu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Rui Yan
School of Artificial Intelligence, Wuhan University, Wuhan, China
4 shared publications - Hongda Sun
Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China
2 shared publications - Bo Du
School of Computer Science, Wuhan University, Wuhan, China
1 shared publication - Cheng Tai
Moqi, Inc., Singapore, Singapore
1 shared publication - Chengwei Wu
Beijing Academy of Artificial Intelligence, Beijing, China
1 shared publication - Cong Li
Lenovo Group Limited, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5454-5463
Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings
Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlying this deficiency. Our motivation stems from an unexpected observation: text embeddings tend to align with frequent but uninformative tokens when projected onto the vocabulary space. We argue that this excessive expression of high-frequency tokens suppresses the model's ability to capture nuanced semantics. To address this, we introduce EmbedFilter, a simple linear transformation designed to refine text embeddings derived from LLMs directly. Specifically, we uncover that the unembedding matrix within LLMs encodes a latent space that is actively writing these frequent tokens into embedding space. By filtering out this subspace, EmbedFilter suppress the influence of high-frequency tokens, thereby enhancing semantic representations. As a compelling byproduct, this enables an inherent dimensionality reduction, lowering index storage and speedup retrieval while fully preserving the refined embedding quality. Our experiments across multiple LLM backbones demonstrate that LLMs equipped with EmbedFilter achieve superior zero-shot downstream performance even with significantly reduced embedding dimensions. We hope our findings provide deeper insights into the mechanisms of LLM-based representations and inspire more principled designs to improve text embeddings training. Our code is available at https://github.com/CentreChen/EmbFilter.
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5454-5463
Open AccessYour UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings
Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlying this deficiency. Our motivation stems from an unexpected observation: text embeddings tend to align with frequent but uninformative tokens when projected onto the vocabulary space. We argue that this excessive expression of high-frequency tokens suppresses the model's ability to capture nuanced semantics. To address this, we introduce EmbedFilter, a simple linear transformation designed to refine text embeddings derived from LLMs directly. Specifically, we uncover that the unembedding matrix within LLMs encodes a latent space that is actively writing these frequent tokens into embedding space. By filtering out this subspace, EmbedFilter suppress the influence of high-frequency tokens, thereby enhancing semantic representations. As a compelling byproduct, this enables an inherent dimensionality reduction, lowering index storage and speedup retrieval while fully preserving the refined embedding quality. Our experiments across multiple LLM backbones demonstrate that LLMs equipped with EmbedFilter achieve superior zero-shot downstream performance even with significantly reduced embedding dimensions. We hope our findings provide deeper insights into the mechanisms of LLM-based representations and inspire more principled designs to improve text embeddings training. Our code is available at https://github.com/CentreChen/EmbFilter.
- Open Access
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 883-893
MobileSteward: Integrating Multiple App-Oriented Agents with Self-Evolution to Automate Cross-App Instructions
- 20 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 883-893
Open AccessMobileSteward: Integrating Multiple App-Oriented Agents with Self-Evolution to Automate Cross-App Instructions
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 504-514
The Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake News
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 504-514
Open AccessThe Truth Becomes Clearer Through Debate! Multi-Agent Systems with Large Language Models Unmask Fake News
- Open Access
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 4372-4382
Harnessing Multi-Role Capabilities of Large Language Models for Open-Domain Question Answering
Open-domain question answering (ODQA) has emerged as a pivotal research spotlight in information systems. Existing methods follow two main paradigms to collect evidence: (1) Theretrieve-then-read paradigm retrieves pertinent documents from an external corpus; and (2) thegenerate-then-read paradigm employs large language models (LLMs) to generate relevant documents. However, neither can fully address multifaceted requirements for evidence. To this end, we propose LLMQA, a generalized framework that formulates the ODQA process into three basic steps: query expansion, document selection, and answer generation, combining the superiority of both retrieval-based and generation-based evidence. Since LLMs exhibit their excellent capabilities to accomplish various tasks, we instruct LLMs to play multiple roles as generators, rerankers, and evaluators within our framework, integrating them to collaborate in the ODQA process. Furthermore, we introduce a novel prompt optimization algorithm to refine role-playing prompts and steer LLMs to produce higher-quality evidence and answers. Extensive experimental results on widely used benchmarks (NQ, WebQ, and TriviaQA) demonstrate that LLMQA achieves the best performance in terms of both answer accuracy and evidence quality, showcasing its potential for advancing ODQA research and applications.
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 4372-4382
Open AccessHarnessing Multi-Role Capabilities of Large Language Models for Open-Domain Question Answering
Open-domain question answering (ODQA) has emerged as a pivotal research spotlight in information systems. Existing methods follow two main paradigms to collect evidence: (1) Theretrieve-then-read paradigm retrieves pertinent documents from an external corpus; and (2) thegenerate-then-read paradigm employs large language models (LLMs) to generate relevant documents. However, neither can fully address multifaceted requirements for evidence. To this end, we propose LLMQA, a generalized framework that formulates the ODQA process into three basic steps: query expansion, document selection, and answer generation, combining the superiority of both retrieval-based and generation-based evidence. Since LLMs exhibit their excellent capabilities to accomplish various tasks, we instruct LLMs to play multiple roles as generators, rerankers, and evaluators within our framework, integrating them to collaborate in the ODQA process. Furthermore, we introduce a novel prompt optimization algorithm to refine role-playing prompts and steer LLMs to produce higher-quality evidence and answers. Extensive experimental results on widely used benchmarks (NQ, WebQ, and TriviaQA) demonstrate that LLMQA achieves the best performance in terms of both answer accuracy and evidence quality, showcasing its potential for advancing ODQA research and applications.