Xinyi Wang

13

Papers

108

Total Citations

Papers (13)

Generalization v.s. Memorization: Tracing Language Models’ Capabilities Back to Pretraining Data

MFABA: A More Faithful and Accelerated Boundary-Based Attribution Method for Deep Neural Networks

Logits DeConfusion with CLIP for Few-Shot Learning

AugRefer: Advancing 3D Visual Grounding via Cross-Modal Augmentation and Spatial Relation-based Referring

GM-MoE: Low-Light Enhancement with Gated-Mechanism Mixture-of-Experts

AffordBot: 3D Fine-grained Embodied Reasoning via Multimodal Large Language Models

CombatVLA: An Efficient Vision-Language-Action Model for Combat Tasks in 3D Action Role-Playing Games

Position: AI/ML Influencers Have a Place in the Academic Process

Iterative Search Attribution for Deep Neural Networks

The Good, The Bad, and Why: Unveiling Emotions in Generative AI

Understanding Reasoning Ability of Language Models From the Perspective of Reasoning Paths Aggregation

Counterfactual Maximum Likelihood Estimation for Training Deep Networks

Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning