Xuebo Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Xuebo Liu has a total of 106 co-authors. Here are the top co-authors by number of shared publications:
- Min Zhang
Harbin Institute of Technology
21 shared publications - Derek Fai Wong
University of Macau
10 shared publications - Dacheng Tao
Nanyang technological university, Singapore, Singapore
7 shared publications - Liang Ding
University of Sydney
6 shared publications - Lidia Sam Chao
University of Macau
6 shared publications - Hexuan Deng
Institute of Computing and Intelligence Harbin Institute of Technology Shenzhen China
4 shared publications
Latest Publications
- Open Access
- 17 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 25  (3) , 1-19
Domain Adaptive Machine Translation with Synthetic Feedback for Large Language Models
Domain-specific machine translation (MT) significantly benefits from large language models (LLMs) due to their strong instruction-following abilities and in-context learning (ICL) capabilities. Appropriate demonstration samples and feedback are essential for helping LLMs refine their translation outputs in real-world applications. However, the scarcity of in-domain samples and professional feedback creates practical limitations. Furthermore, the current ICL paradigm does not offer the fine-grained domain features in addition to parallel translation pairs. To address these challenges, we propose a pipeline that collects in-domain translations from LLMs and generates synthetic, human-like feedback for revising these translations. The translations and their corresponding feedback are stored together to build a demonstration database, with each instance paired with the original in-domain translation and its revision. During online translation, similar in-domain translations can be retrieved as revision demonstrations. This process guides LLMs in iteratively refining their outputs by learning from demonstrations. We evaluate the proposed pipeline using open-source models like Llama3-8B-Instruct and Mistral-7B-Instruct-v0.3, on five domain-specific benchmarks for English-centric, Chinese-centric, and Portuguese-centric translation. The results demonstrate the effectiveness of the pipeline in tailoring in-domain translations and improving translation performance compared to direct translation instructions. Additionally, we discuss the experimental results from the following perspectives: (1) the effectiveness of different in-context retrieval methods; (2) the observed differences across selected domains and language; (3) the quantitative analysis of sentence-level and word-level statistics; and (4) the effect of ICL retrieval database size and decoding parameters.
- 17 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 25  (3) , 1-19
Open AccessDomain Adaptive Machine Translation with Synthetic Feedback for Large Language Models
Domain-specific machine translation (MT) significantly benefits from large language models (LLMs) due to their strong instruction-following abilities and in-context learning (ICL) capabilities. Appropriate demonstration samples and feedback are essential for helping LLMs refine their translation outputs in real-world applications. However, the scarcity of in-domain samples and professional feedback creates practical limitations. Furthermore, the current ICL paradigm does not offer the fine-grained domain features in addition to parallel translation pairs. To address these challenges, we propose a pipeline that collects in-domain translations from LLMs and generates synthetic, human-like feedback for revising these translations. The translations and their corresponding feedback are stored together to build a demonstration database, with each instance paired with the original in-domain translation and its revision. During online translation, similar in-domain translations can be retrieved as revision demonstrations. This process guides LLMs in iteratively refining their outputs by learning from demonstrations. We evaluate the proposed pipeline using open-source models like Llama3-8B-Instruct and Mistral-7B-Instruct-v0.3, on five domain-specific benchmarks for English-centric, Chinese-centric, and Portuguese-centric translation. The results demonstrate the effectiveness of the pipeline in tailoring in-domain translations and improving translation performance compared to direct translation instructions. Additionally, we discuss the experimental results from the following perspectives: (1) the effectiveness of different in-context retrieval methods; (2) the observed differences across selected domains and language; (3) the quantitative analysis of sentence-level and word-level statistics; and (4) the effect of ICL retrieval database size and decoding parameters.
- Open Access
- 12 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 25  (2) , 1-18
Orchestrating Prompt Expertise: Enhancing Knowledge Distillation via Expert-Guided Tuning
Multi-teacher knowledge distillation transfers knowledge from multiple large teacher models to a small student model and has performed well on many downstream tasks. However, when distilling knowledge from multiple teachers, it always suffers from the severe problems of being time-consuming and storage-extensive for multiple teacher models training and inference. We present MoE-KD, a simple but effective framework that produces supervision for training the student model from one single teacher model, which fixes the above problems and improves effectiveness. In the proposed MoE-KD, multiple trainable prompts are used to extract different views of samples from a single pre-trained language model and only a few parameters (prompts) need to be trained and stored. To guarantee the generated supervision signals with increased robustness and correctness, we introduce an uncertainty-based mechanism and a selector module, which routes the input instance to its corresponding teacher. We have also extended MoE KD to lifelong learning scenarios, proposing a lightweight solution for catastrophic forgetting. We conduct experiments on traditional KD scenarios and lifelong learning scenarios. MoE-KD yields improvements up to 1.1% and 140% in accuracy and efficiency in knowledge distillation and 2.8% improvements on average in lifelong learning, compared with the strong baseline methods.
- 12 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Asian and Low-Resource Language Information Processing
- Vol. 25  (2) , 1-18
Open AccessOrchestrating Prompt Expertise: Enhancing Knowledge Distillation via Expert-Guided Tuning
Multi-teacher knowledge distillation transfers knowledge from multiple large teacher models to a small student model and has performed well on many downstream tasks. However, when distilling knowledge from multiple teachers, it always suffers from the severe problems of being time-consuming and storage-extensive for multiple teacher models training and inference. We present MoE-KD, a simple but effective framework that produces supervision for training the student model from one single teacher model, which fixes the above problems and improves effectiveness. In the proposed MoE-KD, multiple trainable prompts are used to extract different views of samples from a single pre-trained language model and only a few parameters (prompts) need to be trained and stored. To guarantee the generated supervision signals with increased robustness and correctness, we introduce an uncertainty-based mechanism and a selector module, which routes the input instance to its corresponding teacher. We have also extended MoE KD to lifelong learning scenarios, proposing a lightweight solution for catastrophic forgetting. We conduct experiments on traditional KD scenarios and lifelong learning scenarios. MoE-KD yields improvements up to 1.1% and 140% in accuracy and efficiency in knowledge distillation and 2.8% improvements on average in lifelong learning, compared with the strong baseline methods.
- Open Access
PACE: Defying the Scaling Hypothesis of Exploration in Iterative Alignment for Mathematical Reasoning
Iterative Direct Preference Optimization has emerged as the state-of-the-art paradigm for aligning Large Language Models on reasoning tasks. Standard implementations (DPO-R1) rely on Best-of-N sampling (e.g., $N \ge 8$) to mine golden trajectories from the distribution tail. In this paper, we challenge this scaling hypothesis and reveal a counter-intuitive phenomenon: in mathematical reasoning, aggressive exploration yields diminishing returns and even catastrophic policy collapse. We theoretically demonstrate that scaling $N$ amplifies verifier noise and induces detrimental distribution shifts. To resolve this, we introduce \textbf{PACE} (Proximal Alignment via Corrective Exploration), which replaces brute-force mining with a generation-based corrective strategy. Operating with a minimal budget ($2<N<3$), PACE synthesizes high-fidelity preference pairs from failed explorations. Empirical evaluations show that PACE outperforms DPO-R1 $(N=16)$ while using only about $1/5$ of the compute, demonstrating superior robustness against reward hacking and label noise.Open AccessPACE: Defying the Scaling Hypothesis of Exploration in Iterative Alignment for Mathematical Reasoning
Iterative Direct Preference Optimization has emerged as the state-of-the-art paradigm for aligning Large Language Models on reasoning tasks. Standard implementations (DPO-R1) rely on Best-of-N sampling (e.g., $N \ge 8$) to mine golden trajectories from the distribution tail. In this paper, we challenge this scaling hypothesis and reveal a counter-intuitive phenomenon: in mathematical reasoning, aggressive exploration yields diminishing returns and even catastrophic policy collapse. We theoretically demonstrate that scaling $N$ amplifies verifier noise and induces detrimental distribution shifts. To resolve this, we introduce \textbf{PACE} (Proximal Alignment via Corrective Exploration), which replaces brute-force mining with a generation-based corrective strategy. Operating with a minimal budget ($2<N<3$), PACE synthesizes high-fidelity preference pairs from failed explorations. Empirical evaluations show that PACE outperforms DPO-R1 $(N=16)$ while using only about $1/5$ of the compute, demonstrating superior robustness against reward hacking and label noise. - Open Access
Stop Rewarding Hallucinated Steps: Faithfulness-Aware Step-Level Reinforcement Learning for Small Reasoning Models
As large language models become smaller and more efficient, small reasoning models (SRMs) are crucial for enabling chain-of-thought (CoT) reasoning in resource-constrained settings. However, they are prone to faithfulness hallucinations, especially in intermediate reasoning steps. Existing mitigation methods based on online reinforcement learning rely on outcome-based rewards or coarse-grained CoT evaluation, which can inadvertently reinforce unfaithful reasoning when the final answer is correct. To address these limitations, we propose Faithfulness-Aware Step-Level Reinforcement Learning (FaithRL), introducing step-level supervision via explicit faithfulness rewards from a process reward model, together with an implicit truncated resampling strategy that generates contrastive signals from faithful prefixes. Experiments across multiple SRMs and Open-Book QA benchmarks demonstrate that FaithRL consistently reduces hallucinations in both the CoT and final answers, leading to more faithful and reliable reasoning. Code is available at https://github.com/Easy195/FaithRL.
Open AccessStop Rewarding Hallucinated Steps: Faithfulness-Aware Step-Level Reinforcement Learning for Small Reasoning Models
As large language models become smaller and more efficient, small reasoning models (SRMs) are crucial for enabling chain-of-thought (CoT) reasoning in resource-constrained settings. However, they are prone to faithfulness hallucinations, especially in intermediate reasoning steps. Existing mitigation methods based on online reinforcement learning rely on outcome-based rewards or coarse-grained CoT evaluation, which can inadvertently reinforce unfaithful reasoning when the final answer is correct. To address these limitations, we propose Faithfulness-Aware Step-Level Reinforcement Learning (FaithRL), introducing step-level supervision via explicit faithfulness rewards from a process reward model, together with an implicit truncated resampling strategy that generates contrastive signals from faithful prefixes. Experiments across multiple SRMs and Open-Book QA benchmarks demonstrate that FaithRL consistently reduces hallucinations in both the CoT and final answers, leading to more faithful and reliable reasoning. Code is available at https://github.com/Easy195/FaithRL.
- Open Access
CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning
Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning (RL) offers a promising alternative by optimizing models through execution-driven rewards without labeled supervision, our preliminary results show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples. We first theoretically analyze showing that branch coverage, sample difficulty, syntactic and functional correctness can be jointly modeled as RL rewards, where optimizing these signals can improve the reliability of unit-test-based verification. Guided by this analysis, we design syntax- and functionality-aware rewards and further propose branch- and sample-difficulty--aware RL using exponential reward shaping and static analysis metrics. With this formulation, CVeDRL achieves state-of-the-art performance with only 0.6B parameters, yielding up to 28.97% higher pass rate and 15.08% higher branch coverage than GPT-3.5, while delivering over $20\times$ faster inference than competitive baselines. Code is available at https://github.com/LIGHTCHASER1/CVeDRL.gitOpen AccessCVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning
Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning (RL) offers a promising alternative by optimizing models through execution-driven rewards without labeled supervision, our preliminary results show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples. We first theoretically analyze showing that branch coverage, sample difficulty, syntactic and functional correctness can be jointly modeled as RL rewards, where optimizing these signals can improve the reliability of unit-test-based verification. Guided by this analysis, we design syntax- and functionality-aware rewards and further propose branch- and sample-difficulty--aware RL using exponential reward shaping and static analysis metrics. With this formulation, CVeDRL achieves state-of-the-art performance with only 0.6B parameters, yielding up to 28.97% higher pass rate and 15.08% higher branch coverage than GPT-3.5, while delivering over $20\times$ faster inference than competitive baselines. Code is available at https://github.com/LIGHTCHASER1/CVeDRL.git - Top Cited Papers
- 27 January 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 337
Exploring and enhancing the transfer of distribution in knowledge distillation for autoregressive language models
- 27 January 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 337
Top Cited PapersExploring and enhancing the transfer of distribution in knowledge distillation for autoregressive language models
- 26 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech and Language Processing
- Vol. 34, 1004-1015
Mitigating Context-Memory Conflicts in LLMs Through Dynamic Cognitive Reconciliation Decoding
Large language models accumulate extensive parametric knowledge through pre-training. However, knowledge conflicts occur when outdated or incorrect parametric knowledge conflicts with external knowledge in the context. Existing methods address knowledge conflicts through contrastive decoding, but in conflict-free scenarios, static approaches disrupt output distribution. Other dynamic decoding methods attempt to measure the degree of conflict but still struggle with complex real-world situations. In this paper, we propose a two-stage decoding method called Dynamic Cognitive Reconciliation Decoding (DCRD), to predict and mitigate context-memory conflicts. DCRD first analyzes the attention map to assess context fidelity and predict potential conflicts. Based on this prediction, the input is directed to one of two decoding paths: (1) greedy decoding, or (2) context fidelity-based dynamic decoding. This design enables DCRD to handle conflicts efficiently while maintaining high accuracy and decoding efficiency in conflict-free cases. Additionally, to simulate scenarios with frequent knowledge updates, we constructed ConflictKG, a knowledge conflict QA benchmark. Experiments on four LLMs across six QA datasets show that DCRD outperforms all baselines, achieving state-of-the-art performance.
- 26 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Audio, Speech and Language Processing
- Vol. 34, 1004-1015
Mitigating Context-Memory Conflicts in LLMs Through Dynamic Cognitive Reconciliation Decoding
Large language models accumulate extensive parametric knowledge through pre-training. However, knowledge conflicts occur when outdated or incorrect parametric knowledge conflicts with external knowledge in the context. Existing methods address knowledge conflicts through contrastive decoding, but in conflict-free scenarios, static approaches disrupt output distribution. Other dynamic decoding methods attempt to measure the degree of conflict but still struggle with complex real-world situations. In this paper, we propose a two-stage decoding method called Dynamic Cognitive Reconciliation Decoding (DCRD), to predict and mitigate context-memory conflicts. DCRD first analyzes the attention map to assess context fidelity and predict potential conflicts. Based on this prediction, the input is directed to one of two decoding paths: (1) greedy decoding, or (2) context fidelity-based dynamic decoding. This design enables DCRD to handle conflicts efficiently while maintaining high accuracy and decoding efficiency in conflict-free cases. Additionally, to simulate scenarios with frequent knowledge updates, we constructed ConflictKG, a knowledge conflict QA benchmark. Experiments on four LLMs across six QA datasets show that DCRD outperforms all baselines, achieving state-of-the-art performance.
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28  (15209210) , 1677-1688
Exploiting Multimodal Knowledge Graph for Multimodal Machine Translation
A neural Multimodal Machine Translation (MMT) system utilizes multimodal information, particularly images, to enhance traditional text-only models and achieve superior performance. However, the effectiveness of MMT heavily depends on the availability of extensive collections of bilingual parallel sentence pairs and manually annotated images, which poses a challenge due to the scarcity of such pairs. To address this issue, we propose incorporating the Multimodal Knowledge Graph (MMKG) for data augmentation in MMT. By utilizing MMKG as an additional source of knowledge, we can overcome the limitations of existing sentence-image pairings. This allows us to expand the original parallel corpus and generate corresponding images, creating new synthetic data pairs that facilitate effective data augmentation. Experiments conducted on two translation datasets, Multi30k and IKEA, demonstrate that the proposed MMKG enhancement method significantly improves performance across multiple baseline methods, ultimately outperforming all baseline approaches. Additionally, experiments under low-resource conditions reveal that our method achieves exceptional enhancement effects in low-resource corpora, surpassing other data augmentation baseline methods. These results indicate the efficacy and potential of the proposed method for enhancing the performance of multimodal models across diverse datasets.
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28  (15209210) , 1677-1688
Exploiting Multimodal Knowledge Graph for Multimodal Machine Translation
A neural Multimodal Machine Translation (MMT) system utilizes multimodal information, particularly images, to enhance traditional text-only models and achieve superior performance. However, the effectiveness of MMT heavily depends on the availability of extensive collections of bilingual parallel sentence pairs and manually annotated images, which poses a challenge due to the scarcity of such pairs. To address this issue, we propose incorporating the Multimodal Knowledge Graph (MMKG) for data augmentation in MMT. By utilizing MMKG as an additional source of knowledge, we can overcome the limitations of existing sentence-image pairings. This allows us to expand the original parallel corpus and generate corresponding images, creating new synthetic data pairs that facilitate effective data augmentation. Experiments conducted on two translation datasets, Multi30k and IKEA, demonstrate that the proposed MMKG enhancement method significantly improves performance across multiple baseline methods, ultimately outperforming all baseline approaches. Additionally, experiments under low-resource conditions reveal that our method achieves exceptional enhancement effects in low-resource corpora, surpassing other data augmentation baseline methods. These results indicate the efficacy and potential of the proposed method for enhancing the performance of multimodal models across diverse datasets.
- Open Access
- 3 June 2025
- Published by MDPI AG in Mathematics
- Vol. 13  (11) , 1874
Noise Improves Multimodal Machine Translation: Rethinking the Role of Visual Context
Multimodal Machine Translation (MMT) has long been assumed to outperform traditional text-only MT by leveraging visual information. However, recent studies challenge this assumption, showing that MMT models perform similarly even when tested without images or with mismatched images. This raises fundamental questions about the actual utility of visual information in MMT, which this work aims to investigate. We first revisit commonly used image-must and image-free MMT approaches, identifying that suboptimal performance may stem from insufficiently robust baseline models. To further examine the role of visual information, we propose a novel visual type regularization method and introduce two probing tasks—Visual Contribution Probing and Modality Relationship Probing—to analyze whether and how visual features influence a strong MMT model. Surprisingly, our findings on a mainstream dataset indicate that the gains from visual information are marginal. We attribute this improvement primarily to a regularization effect, which can be replicated using random noise. Our results suggest that the MMT community should critically re-evaluate baseline models, evaluation metrics, and dataset design to advance multimodal learning meaningfully.
- 3 June 2025
- Published by MDPI AG in Mathematics
- Vol. 13  (11) , 1874
Open AccessNoise Improves Multimodal Machine Translation: Rethinking the Role of Visual Context
Multimodal Machine Translation (MMT) has long been assumed to outperform traditional text-only MT by leveraging visual information. However, recent studies challenge this assumption, showing that MMT models perform similarly even when tested without images or with mismatched images. This raises fundamental questions about the actual utility of visual information in MMT, which this work aims to investigate. We first revisit commonly used image-must and image-free MMT approaches, identifying that suboptimal performance may stem from insufficiently robust baseline models. To further examine the role of visual information, we propose a novel visual type regularization method and introduce two probing tasks—Visual Contribution Probing and Modality Relationship Probing—to analyze whether and how visual features influence a strong MMT model. Surprisingly, our findings on a mainstream dataset indicate that the gains from visual information are marginal. We attribute this improvement primarily to a regularization effect, which can be replicated using random noise. Our results suggest that the MMT community should critically re-evaluate baseline models, evaluation metrics, and dataset design to advance multimodal learning meaningfully.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 1086-1098
Weight-Aware Activation Sparsity with Constrained Bayesian Optimization Scheduling for Large Language Models
Ming Wang, Miao Zhang, Xuebo Liu, Liqiang Nie. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 1086-1098
Open AccessWeight-Aware Activation Sparsity with Constrained Bayesian Optimization Scheduling for Large Language Models
Ming Wang, Miao Zhang, Xuebo Liu, Liqiang Nie. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.