PPaperPicks

Mengdi Wang

Princeton University, Center for Statistics and Machine Learning, Department of Electrical and Computer Engineering, NJ, USA

44 papers at tracked venues · 38 at CORE A* · active 2024–2026

DBLP profile ↗ORCID 0000-0002-2101-9507 ↗Homepage ↗

Venues

NeurIPS ×16ICML ×9ICLR ×8EMNLP ×3AAAI ×2AISTATS ×1CVPR ×1EACL ×1ICCV ×1IJCAI ×1SIGKDD ×1

Frequent coauthors

Souradip Chakraborty
DBLP profile ↗ORCID search ↗
×4
Jiahao Qiu
DBLP profile ↗ORCID search ↗
×3
Soumya Suvra Ghosal
DBLP profile ↗ORCID search ↗
×2
Zaixi Zhang
DBLP profile ↗ORCID search ↗
×2
Jiacheng Guo
DBLP profile ↗ORCID search ↗
×2
Yingqing Guo
DBLP profile ↗ORCID search ↗
×2
Zihao Li
DBLP profile ↗ORCID search ↗
×2
James Beetham
DBLP profile ↗ORCID search ↗
×1
Hui Yuan
DBLP profile ↗ORCID search ↗
×1
Quan Xiao
DBLP profile ↗ORCID search ↗
×1
Yinjie Wang
DBLP profile ↗ORCID search ↗
×1
Jonathan Light
DBLP profile ↗ORCID search ↗
×1

Papers

  1. Jailbreaks as Inference-Time Alignment: A Framework for Understanding Safety Failures in LLMs
    EACL 2026 ·
    James Beetham
    DBLP profile ↗ORCID search ↗
  2. A Common Pitfall of Margin-based Language Model Alignment: Gradient Entanglement
    ICLR 2025 ·
    Hui Yuan
    DBLP profile ↗ORCID search ↗
  3. A First-order Generative Bilevel Optimization Framework for Diffusion Models
    ICML 2025 ·
    Quan Xiao
    DBLP profile ↗ORCID search ↗
  4. Co-Evolving LLM Coder and Unit Tester via Reinforcement Learning
    NeurIPS 2025 ·
    Yinjie Wang
    DBLP profile ↗ORCID search ↗
  5. Collab: Controlled Decoding using Mixture of Agents for LLM Alignment
    ICLR 2025 ·
    Souradip Chakraborty
    DBLP profile ↗ORCID search ↗
  6. DISC: Dynamic Decomposition Improves LLM Inference Scaling
    NeurIPS 2025 ·
    Jonathan Light
    DBLP profile ↗ORCID search ↗
  7. Deep Reinforcement Learning for Efficient and Fair Allocation of Healthcare Resources
    IJCAI 2025 ·
    Yikuan Li
    DBLP profile ↗ORCID search ↗
  8. Diffusion Transformer Captures Spatial-Temporal Dependencies: A Theory for Gaussian Process Data
    ICLR 2025 ·
    Hengyu Fu
    DBLP profile ↗ORCID search ↗
  9. Does Thinking More Always Help? Mirage of Test-Time Scaling in Reasoning Models
    NeurIPS 2025 ·
    Soumya Suvra Ghosal
    DBLP profile ↗ORCID search ↗
  10. Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language Models
    ICML 2025 ·
    Yukang Yang
    DBLP profile ↗ORCID search ↗
  11. EmoAgent: Assessing and Safeguarding Human-AI Interaction for Mental Health Safety
    EMNLP 2025 ·
    Jiahao Qiu
    DBLP profile ↗ORCID search ↗
  12. Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment
    CVPR 2025 ·
    Soumya Suvra Ghosal
    DBLP profile ↗ORCID search ↗
  13. IterComp: Iterative Composition-Aware Feedback Learning from Model Gallery for Text-to-Image Generation
    ICLR 2025 ·
    Xinchen Zhang
    DBLP profile ↗ORCID search ↗
  14. MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations
    ICML 2025 ·
    Kaixuan Huang
    DBLP profile ↗ORCID search ↗
  15. MMaDA: Multimodal Large Diffusion Language Models
    NeurIPS 2025 ·
    Ling Yang
    DBLP profile ↗ORCID search ↗
  16. Preacher: Paper-to-Video Agentic System
    ICCV 2025 ·
    Jingwei Liu
    DBLP profile ↗ORCID search ↗
  17. ReasonFlux-PRM: Trajectory-Aware PRMs for Long Chain-of-Thought Reasoning in LLMs
    NeurIPS 2025 ·
    Jiaru Zou
    DBLP profile ↗ORCID search ↗
  18. Rectified Diffusion: Straightness Is Not Your Need in Rectified Flow
    ICLR 2025 ·
    Fu-Yun Wang
    DBLP profile ↗ORCID search ↗
  19. Securing the Language of Life: Inheritable Watermarks from DNA Language Models to Proteins
    NeurIPS 2025 ·
    Zaixi Zhang
    DBLP profile ↗ORCID search ↗
  20. Temporal Consistency for LLM Reasoning Process Error Identification
    EMNLP 2025 ·
    Jiacheng Guo
    DBLP profile ↗ORCID search ↗
  21. Towards Understanding Text Hallucination of Diffusion Models via Local Generation Bias
    ICLR 2025 ·
    Rui Lu
    DBLP profile ↗ORCID search ↗
  22. Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models
    NeurIPS 2025 ·
    Yingqing Guo
    DBLP profile ↗ORCID search ↗
  23. TreeBoN: Enhancing Inference-Time Alignment with Speculative Tree-Search and Best-of-N Sampling
    EMNLP 2025 ·
    Jiahao Qiu
    DBLP profile ↗ORCID search ↗
  24. A Theoretical Perspective for Speculative Decoding Algorithm
    NeurIPS 2024 ·
    Ming Yin
    DBLP profile ↗ORCID search ↗
  25. Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
    ICML 2024 ·
    Boyi Wei
    DBLP profile ↗ORCID search ↗
  26. Conversational Dueling Bandits in Generalized Linear Models
    SIGKDD 2024 ·
    Shuhua Yang
    DBLP profile ↗ORCID search ↗
  27. Fast Best-of-N Decoding via Speculative Rejection
    NeurIPS 2024 ·
    Hanshi Sun
    DBLP profile ↗ORCID search ↗
  28. FlexSBDD: Structure-Based Drug Design with Flexible Protein Modeling
    NeurIPS 2024 ·
    Zaixi Zhang
    DBLP profile ↗ORCID search ↗
  29. Global Convergence in Training Large-Scale Transformers
    NeurIPS 2024 ·
    Cheng Gao
    DBLP profile ↗ORCID search ↗
  30. Gradient Guidance for Diffusion Models: An Optimization Perspective
    NeurIPS 2024 ·
    Yingqing Guo
    DBLP profile ↗ORCID search ↗
  31. Information-Directed Pessimism for Offline Reinforcement Learning
    ICML 2024 ·
    Alec Koppel
    DBLP profile ↗ORCID search ↗
  32. Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective
    ICML 2024 ·
    Lei Zhao
    DBLP profile ↗ORCID search ↗
  33. MaxMin-RLHF: Alignment with Diverse Human Preferences
    ICML 2024 ·
    Souradip Chakraborty
    DBLP profile ↗ORCID search ↗
  34. Nonparametric Classification on Low Dimensional Manifolds using Overparameterized Convolutional Residual Networks
    NeurIPS 2024 ·
    Zixuan Zhang
    DBLP profile ↗ORCID search ↗
  35. Offline Multitask Representation Learning for Reinforcement Learning
    NeurIPS 2024 ·
    Haque Ishfaq
    DBLP profile ↗ORCID search ↗
  36. One-Layer Transformer Provably Learns One-Nearest Neighbor In Context
    NeurIPS 2024 ·
    Zihao Li
    DBLP profile ↗ORCID search ↗
  37. PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback
    ICLR 2024 ·
    Souradip Chakraborty
    DBLP profile ↗ORCID search ↗
  38. Policy Evaluation for Reinforcement Learning from Human Feedback: A Sample Complexity Analysis
    AISTATS 2024 ·
    Zihao Li
    DBLP profile ↗ORCID search ↗
  39. Sample-Efficient Learning of POMDPs with Multiple Observations In Hindsight
    ICLR 2024 ·
    Jiacheng Guo
    DBLP profile ↗ORCID search ↗
  40. Theoretical insights for diffusion guidance: A case study for Gaussian mixture models
    ICML 2024 ·
    Yuchen Wu
    DBLP profile ↗ORCID search ↗
  41. Theory of Consistency Diffusion Models: Distribution Estimation Meets Fast Sampling
    ICML 2024 ·
    Zehao Dou
    DBLP profile ↗ORCID search ↗
  42. Transfer Q-star : Principled Decoding for LLM Alignment
    NeurIPS 2024 ·
    Souradip Chakraborty
    DBLP profile ↗ORCID search ↗
  43. Tree Search-Based Evolutionary Bandits for Protein Sequence Optimization
    AAAI 2024 ·
    Jiahao Qiu
    DBLP profile ↗ORCID search ↗
  44. Visual Adversarial Examples Jailbreak Aligned Large Language Models
    AAAI 2024 ·
    Xiangyu Qi
    DBLP profile ↗ORCID search ↗