Mengnan Du

KnowThyself: An Agentic Assistant for LLM Interpretability AAAI 2026 SAGE: An Agentic Explainer Framework for Interpreting SAE Features in Language Models EACL 2026 FaithLM: Towards Faithful Explanations for Large Language Models EACL 2026 LLM Agents in Law: Taxonomy, Applications, and Challenges ACL 2026 FinChart-Bench: Benchmarking Financial Chart Comprehension in Vision-Language Models ACL 2026 FinCall-Surprise: A Large Scale Multi-modal Benchmark for Earning Surprise Prediction ACL 2026 AdaJudge: Adaptive Multi-Perspective Judging for Reward Modeling ACL 2026 Fine-Grained Interpretation of Political Opinions in Large Language Models AAAI 2026 DeepSieve: Information Sieving via LLM-as-a-Knowledge-Router EACL 2026 Denoising Concept Vectors with Sparse Autoencoders for Improved Language Model Steering EACL 2026 Exploring Multilingual Probing in Large Language Models: A Cross-Language Analysis ACL 2025 Improving LLM Reasoning through Interpretable Role-Playing Steering EMNLP 2025 A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models EMNLP 2025 Language Ranker: A Metric for Quantifying LLM Performance Across High and Low-Resource Languages AAAI 2025 Comparative Analysis of Demonstration Selection Algorithms for In-Context Learning in Large Language Models (Student Abstract) AAAI 2025 Exploring Concept Depth: How Large Language Models Acquire Knowledge and Concept at Different Layers? COLING 2025 Invisible Backdoor Attack against Self-supervised Learning CVPR 2025 Large Vision-Language Model Alignment and Misalignment: A Survey Through the Lens of Explainability EMNLP 2025 From Commands to Prompts: LLM-based Semantic File System for AIOS ICLR 2025 Beyond Single Concept Vector: Modeling Concept Subspace in LLMs with Gaussian Distribution ICLR 2025 ContractEval: Benchmarking LLMs for Clause-Level Legal Risk Identification in Commercial Contracts EMNLP 2025 Beyond Input Activations: Identifying Influential Latents by Gradient Sparse Autoencoders EMNLP 2025 SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models EMNLP 2025 Massive Values in Self-Attention Modules are the Key to Contextual Knowledge Understanding ICML 2025 Concept-Centric Token Interpretation for Vector-Quantized Generative Models ICML 2025 Data-centric NLP Backdoor Defense from the Lens of Memorization NAACL 2025 Feature Extraction and Steering for Enhanced Chain-of-Thought Reasoning in Language Models EMNLP 2025 Strategic Demonstration Selection for Improved Fairness in LLM In-Context Learning EMNLP 2024 The Impact of Reasoning Step Length on Large Language Models ACL 2024 Data-Centric Explainable Debiasing for Improving Fairness in Pre-trained Language Models ACL 2024 Knowledge Graph Large Language Model (KG-LLM) for Link Prediction ACML 2024 DataFrame QA: A Universal LLM Framework on DataFrame Question Answering Without Data Exposure ACML 2024 Mitigating Shortcuts in Language Models with Soft Label Encoding COLING 2024 Unveiling Project-Specific Bias in Neural Code Models COLING 2024 Explaining Time Series via Contrastive and Locally Sparse Perturbations ICLR 2024 TVE: Learning Meta-attribution for Transferable Vision Explainer ICML 2024 Secure Your Model: An Effective Key Prompt Protection Mechanism for Large Language Models NAACL 2024 $\mathcal{M}^4$: A Unified XAI Benchmark for Faithfulness Evaluation of Feature Attribution Methods across Metrics, Modalities and Models NIPS 2023 Fairness via Group Contribution Matching IJCAI 2023 FAIRER: Fairness as Decision Rationale Alignment ICML 2023 Robustness Challenges in Model Distillation and Pruning for Natural Language Understanding EACL 2023 Prompt Tuning Pushes Farther, Contrastive Learning Pulls Closer: A Two-Stage Approach to Mitigate Social Biases ACL 2023 Black-box Backdoor Defense via Zero-shot Image Purification NIPS 2023 Towards Debiasing DNN Models from Spurious Feature Influence AAAI 2022 Accelerating Shapley Explanation via Contributive Cooperator Selection ICML 2022 DEGREE: Decomposition Based Explanation for Graph Neural Networks ICLR 2022 A Unified Taylor Framework for Revisiting Attribution Methods AAAI 2021 Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU models NAACL 2021 Fairness via Representation Neutralization NIPS 2021