AI.info

Xi Chen

Explore Xi Chen on AI.info.

• PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives
• OmniEdit-Bench: A Comprehensive Benchmark for Instruction-based Video Editing
• LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories
• Lost in the Hype: Revealing and Dissecting the Performance Degradation of Medical Multimodal Large Language Models in Image Classification
• Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching
• ShapeCond: Fast Shapelet-Guided Dataset Condensation for Time Series Classification
• MobileManiBench: Simplifying Model Verification for Mobile Manipulation
• Reward-free Alignment for Conflicting Objectives
• LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization
• All That Glisters Is Not Gold: A Benchmark for Reference-Free Counterfactual Financial Misinformation Detection
• Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
• Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward
• Ambiguity Awareness Optimization: Towards Semantic Disambiguation for Direct Preference Optimization
• Seg-VAR: Image Segmentation with Visual Autoregressive Modeling
• ARCHE: A Novel Task to Evaluate LLMs on Latent Reasoning Chain Extraction
• SPAN: Spatial-Projection Alignment for Monocular 3D Object Detection
• Weight Decay may matter more than muP for Learning Rate Transfer in Practice
• Aligning Deep Implicit Preferences by Learning to Reason Defensively