Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Xueqi ChengSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works137 from public data
- Text Matching as Image Recognition579
This model, which resembles the compositional hierarchies of patterns in image recognition, can successfully identify salient signals such as n-gram and n-term matchings and demonstrates its superiority against the baselines.
- A Deep Look into neural ranking models for information retrieval395
A deep look into the neural ranking models from different dimensions is taken to analyze their underlying assumptions, major design principles, and learning strategies to obtain a comprehensive empirical understanding of the existing techniques.
- A Deep Architecture for Semantic Matching with Multiple Positional Sentence Representations359
This work presents a new deep architecture to match two sentences with multiple positional sentence representations, generated by a bidirectional long short term memory (Bi-LSTM).
- Bias and Unfairness in Information Retrieval Systems: New Challenges in the LLM Era247
A comprehensive survey of existing works on emerging and pressing bias and unfairness issues in IR systems when the integration of LLMs is presented, and systematically delve into the specific bias and unfairness issues arising from three critical stages of LLMs integration into IR systems.
- SetRank169
A neural learning-to-rank model called SetRank is proposed which directly learns a permutation-invariant ranking model defined on document sets of any size and Experimental results showed that the SetRank significantly outperformed the baselines include the traditional learning- to-rank models and state-of-the-art Neural IR models.
- HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data167
This work presents a novel hallucination detection and elimination framework, HalluciDoctor, based on the cross-checking paradigm, and uses this framework to identify and eliminate hallucinations in the training data automatically, and indicates that spurious correlations arising from long-tail object cooccurrences contribute to hallucinations.
- 131
- ReCoSa: Detecting the Relevant Contexts with Self-Attention for Multi-turn Dialogue Generation129
Experimental results on both Chinese customer services dataset and English Ubuntu dialogue dataset show that ReCoSa significantly outperforms baseline models, in terms of both metric-based and human evaluations.
- Search-in-the-Chain: Interactively Enhancing Large Language Models with Search for Knowledge-intensive Tasks122
This paper proposes a novel framework named Search-in-the-Chain (SearChain) for the interaction between LLM and IR to solve the challenges and shows that SearChain outperforms state-of-the-art baselines on complex knowledge-intensive tasks including multi-hop Q&A, slot filling, fact checking, and long-form Q&A.
- Can Small Language Models be Good Reasoners for Sequential Recommendation?88
A novel Step-by-step knowLedge knowLedge fraMework for recommendation (SLIM) is proposed, paving a promising path for sequential recommenders to enjoy the exceptional reasoning capabilities of LLMs in a "slim" (i.e. resource-efficient) manner.
- LLMDet: A Third Party Large Language Models Generated Text Detection Tool88
LLMDet is a model-specific, secure, efficient, and extendable detection tool that can source text from specific LLMs, such as GPT-2, OPT, LLaMA, and others, that can effortlessly extend its detection capabilities to a new open-source model.
- 83
- Beyond Factoid QA: Effective Methods for Non-factoid Answer Sentence Retrieval79
This work introduces the answer sentence retrieval task for non-factoid Web queries, and investigates how this task can be effectively solved under a learning to rank framework, and designs two types of features, namely semantic and context features, beyond traditional text matching features.
- Neural Retrievers are Biased Towards LLM-Generated Content76
A quantitative evaluation of IR models in scenarios where both human-written and LLM-generated texts are involved indicates that neural retrieval models tend to rank LLM-generated documents higher, and proposes a plug-and-play debiased constraint for the optimization objective to mitigate the source bias.
- Modeling Topical Relevance for Multi-Turn Dialogue Generation53
A new model, named STAR-BTM, is proposed, which significantly outperforms several state-of-the-art methods, in terms of both metric-based and human evaluations.
- Adaptive Information Seeking for Open-Domain Question Answering44
A novel adaptive information-seeking strategy for open-domain question answering, namely AISO is proposed, which outperforms all baseline methods with predefined strategies in terms of both retrieval and answer evaluations.
- HAS-QA: Hierarchical Answer Spans Model for Open-Domain Question Answering43
A new probabilistic formulation of OpenQA is proposed, based on a three-level hierarchical structure, i.e.,~the question level, the paragraph level and the answer span level, and a Hierarchical Answer Spans Model (HAS-QA) is designed to capture each probability.
- LightPROF: A Lightweight Reasoning Framework for Large Language Model on Knowledge Graph39
A novel Lightweight and efficient Prompt learning-ReasOning Framework for KGQA (LightPROF), which leverages the full potential of LLMs to tackle complex reasoning tasks in a parameter-efficient manner and shows significant advantages in terms of input token count and reasoning time.
- 38
- Transductive Learning for Unsupervised Text Style Transfer33
This paper proposes a novel transductive learning approach, based on a retrieval-based context-aware style representation, which involves top-K relevant sentences in the target style in the transfer process and outperforms several strong baselines.
- Transformation Driven Visual Reasoning32
A novel transformation driven visual reasoning task, where the target is to infer the corresponding single-step or multi-step transformation, represented as a triplet or a sequence of triplets, respectively, which will boost the development of machine visual reasoning.
- A Survey of Generative Search and Recommendation in the Era of Large Language Models31
This paper provides a comprehensive survey of the emerging paradigm in information systems and summarize the developments in generative search and recommendation from a unified perspective, and abstracts a unified framework for the generative paradigm.
- Match-Ignition28
The main idea is to plug the well-known PageRank algorithm into the Transformer, to identify and filter both sentence and word level noisy information in the matching process, and to show that Match-Ignition efficiently captures important sentences and words, to facilitate the long-form text matching process.
- Invisible Relevance Bias: Text-Image Retrieval Models Prefer AI-Generated Images27
The proposed debiasing method is applied to retroactively identify the causes of the invisible relevance bias, revealing that the AI-generated images induce the image encoder to embed additional information into their representation, which makes the retriever estimate a higher relevance score.
- SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement26
SkillAttack is introduced, a red-teaming framework that dynamically verifies skill vulnerability exploitability through adversarial prompting, and outperforms all baselines by a wide margin, revealing that even well-intended skills pose serious security risks under realistic agent interactions.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.