Min‐Yen Kan lists you as their PhD student. Claim this profile to confirm it and keep the rest of your record right.
Claim this profileAcademic lineage
Traced back 35 generations →Advisors
Works52 from public data
- Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies307
A comprehensive review of techniques leveraging automated feedback to make LLM-based solutions more practical and deployable with minimal human feedback, including training-time, generation- time, and post-hoc correction.
- Automatically Correcting Large Language Models: Surveying the Landscape of Diverse Automated Correction Strategies191
An exhaustive review of the recent advances in correcting LLMs with automated feedback, categorizing them into training-time, generation-time, and post-hoc approaches and identifying potential challenges and future directions in this emerging field.
- 173
- Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment163
An attributed value encoder is utilized and the KG is partitioned into subgraphs to model the various types of attribute triples efficiently and achieves significant improvements in cross-lingual and monolingual datasets.
- KQA Pro: A Dataset with Explicit Compositional Programs for Complex Question Answering over Knowledge Base154
Experimental results show that state-of-the-art KBQA methods cannot achieve promising results on KQA Pro as on current datasets, which suggests that K QA Pro is challenging and Complex KBZA requires further research efforts.
- Prerequisite Relation Learning for Concepts in MOOCs137
This work proposes a representation learning-based method for learning latent representations of course concepts, and investigates how different features capture the prerequisite relations between concepts.
- Semantic Graphs for Generating Deep Questions112
A novel framework that first constructs a semantic-level graph for the input document and then encodes the semantic graph by introducing an attention-based GGNN (Att-GGNN) to perform joint training of content selection and question decoding is proposed.
- Expertise Style Transfer: A New Task Towards Better Communication between Experts and Laymen87
A new task of expertise style transfer is proposed and a manually annotated dataset is contributed with the goal of alleviating cognitive biases and improving the accuracy and expertise level of laymen descriptions using simple words.
- 82
- Zero-shot Fact Verification by Claim Generation70
QACG, a framework for training a robust fact verification model by using automatically generated claims that can be supported, refuted, or unverifiable from evidence from Wikipedia, is developed.
- Unsupervised Multi-hop Question Answering by Question Generation68
This work proposes MQA-QG, an unsupervised framework that can generate human-like multi-hop training data from both homogeneous and heterogeneous data sources and shows that pretraining the QA system with the generated data would greatly reduce the demand for human-annotated training data.
- Zero-Shot Ingredient Recognition by Multi-Relational Graph Convolutional Network68
Multi-relational GCN (graph convolutional network) is introduced that integrates ingredient hierarchy, attribute as well as co-occurrence for zero-shot ingredient recognition and sheds light on zero- shot ingredients recognition.
- Course Concept Extraction in MOOCs via Embedding-Based Graph Propagation60
This paper proposes to learn latent representations for candidate concepts via an embedding-based method and develops a graph-based propagation algorithm to rank the candidate concepts based on the learned representations.
- AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge53
This paper proposes AntiLeak-Bench, an automated anti-leakage benchmarking framework that significantly reduces the cost of benchmark maintenance to accommodate emerging LLMs and highlights that data contamination likely exists before LLMs' cutoff time.
- Attacking Open-domain Question Answering by Injecting Misinformation53
The necessity of building a misinformation-aware QA system that integrates question-answering and misinformation detection in a joint fashion is discussed, with experiments showing that QA models are vulnerable to even small amounts of evidence contamination brought by misinformation.
- Towards Verifiable Generation: A Benchmark for Knowledge-aware Language Model Attribution35
A new task of Knowledge-aware Language Model Attribution (KaLMA) is defined that improves upon three core concerns with conventional attributed LMs, and a new ``Conscious Incompetence" setting is proposed considering the incomplete knowledge repository.
- Exploring Question-Specific Rewards for Generating Deep Questions35
This work designs three different rewards that target to improve the fluency, relevance, and answerability of generated questions and finds that optimizing on question-specific rewards generally leads to better performance in automatic evaluation metrics.
- CausalEval: Towards Better Causal Reasoning in Language Models30
A comprehensive review of research aimed at enhancing LMs for causal reasoning, coupled with an empirical evaluation of current models and methods, and assess the performance of current LMs and various enhancement methods on a range of causal reasoning tasks.
- FOLLOWUPQG: Towards Information-Seeking Follow-up Question Generation25
This work constructs FOLLOWUPQG, a dataset of over 3K real-world tuples collected from a Reddit forum providing layman-friendly explanations for open-ended questions, and evaluates current question generation models on their efficacy for generating follow-up questions.
- 24
- 24
- KQA Pro: A Large Diagnostic Dataset for Complex Question Answering over Knowledge Base.18
This work introduces KQA Pro, a large-scale dataset for Complex KBQA, and generates questions, SPARQLs, and functional programs with recursive templates and paraphrase the questions by crowdsourcing, giving rise to around 120K diverse instances.
- How do Transformers Learn Implicit Reasoning?17
This paper studies how implicit multi-hop reasoning emerges by training transformers from scratch in a controlled symbolic environment, revealing a three-stage developmental trajectory: early memorization, followed by in-distribution generalization, and eventually cross-distribution generalization.
- Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models14
The results show that LLMs tend to suppress their PK when contextual information is available, even when it is complementary or irrelevant, revealing a key vulnerability in LLMs, raising concerns about their reliability in knowledge-intensive tasks.
- Resource Mention Extraction for MOOC Discussion Forums13
This work proposes the novel problem of learning resource mention identification in MOOC forums, i.e., to identify resource mentions in discussions and classify them into pre-defined resource types, and identifies two major challenges that hinder the applications of sequence tagging models to the task.
- Hashtag-Guided Low-Resource Tweet Classification11
Experiments show that HashTation achieves significant improvements on seven low-resource tweet classification tasks, in which only a limited amount of training data is provided, showing that automatically enriching tweets with model-generated hashtags could significantly reduce the demand for large-scale human-labeled data.
- Long Context vs. RAG: Strategies for Processing Long Documents in LLMs10
This half-day tutorial offers a unified, beginner-friendly introduction to Retrieval-Augmented Generation (RAG) and Long Context methods, equipping IR researchers and practitioners with actionable guidelines for processing long documents in LLMs.
- Towards Temporal-Aware Multi-Modal Retrieval Augemented Generation in Finance6
This work introduces FINTMMBench, the first comprehensive benchmark for evaluating temporal-aware multi-modal Retrieval-Augmented Generation (RAG) systems in finance, and proposes a novel TMMHybridRAG method, which first leverages a multi-modal LLM to convert data from other modalities into textual format and then incorporates temporal information in each node when constructing graphs and dense indexes.
- 6
- Ingredient-enriched Recipe Generation from Cooking Videos5
A framework that introduces ingredient recognition module which uses the copy mechanism to fuse the predicted ingredient information into the generated sentence is proposed which achieves the promising performances on both YouCookII and Cooking-COIN datasets.
- 5
- COrAL: Order-Agnostic Language Modeling for Efficient Iterative Refinement3
Context-Wise Order-Agnostic Language Modeling (COrAL) is proposed, which incorporates iterative refinement directly into the LLM architecture while maintaining computational efficiency and introduces sliding blockwise order-agnostic decoding, which performs multi-token forward prediction and backward reconstruction within context windows.
- Domain Specific Cross-Lingual Knowledge Linking Based on Similarity Flooding3
This paper proposes an unsupervised framework for cross-lingual knowledge linking on specific domains, different from traditional methods, and evaluates the proposed method on two knowledge linking tasks to find English-Chinese knowledge links.
- RiMOM results for OAEI 2016.3
The overall framework of RiMOM (RiMOM), which is designed for large-scale and cross-lingual instance matching task specially, is described and an extended version, which provides support for cross-lingual instance matching in a supervised or an unsupervised way is proposed.
- Towards a Mechanistic Understanding of Propositional Logical Reasoning in Large Language Models2
These findings show that pretrained LLMs solve propositional reasoning through a structured, layer-organized process that recurs across models, rule categories, and reasoning hops.
- Doolittle: Benchmarks and Corpora for Academic Writing Formalization2
This work forms this language refinement task as a formal text style transfer task which transfers informal-academic text to formal-academic and contributes a large-scale non-parallel dataset for this purpose.
- A Hybrid Approach for Detecting Prerequisite Relations in Multi-Modal Food Recipes2
This work proposes a simple-yet-effective two-stage framework to automatically construct the prerequisite tree for a recipe by utilizing a trained classifier to detect pairwise prerequisite relations that fuses multi-modal features as input.
- 2
- LEDOM: Reverse Language Model1
This work trains LEDOM, an open-source purely reverse autoregressive language model, and finds it develops capabilities distinct from forward models, including abductive inference, question synthesis, and natural resolution of the reversal curse.
- 1
- 1
- Boosting to Build a Large-Scale Cross-Lingual Ontology1
This paper proposes a cross-lingually boosting approach to iteratively reinforce the performance of ontology building and instance matching, and outputs an ontology containing over 3,520,000 English instances, 800,000 Chinese instances, and over 150,000 cross-lingsual instance alignments.
- –
- –
- –
- –
- HOLMES: Evaluating Higher-Order Logical Reasoning in LLMs–
HOLMES (Higher-Order Logic Meets real-world Explainable Symbolic reasoning), the first real-world benchmark for higher-order symbolic reasoning in LLMs, is introduced, revealing that high final-answer accuracy can mask shortcut reasoning in conflict-resolution settings, while performance drops sharply under scope-conditioned and compositional reasoning.
- –
- –
- –
- –
- –
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.