Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMs

Documenti analoghi

• View-Aware Semantic Alignment for Aerial-Ground Person Re-Identification
  di: Zhang, Quan, et al.
  Pubblicazione: (2026)
• Thinking Before Matching: A Reinforcement Reasoning Paradigm Towards General Person Re-Identification
  di: Zhang, Quan, et al.
  Pubblicazione: (2026)
• From GPT-4 to Gemini and Beyond: Assessing the Landscape of MLLMs on Generalizability, Trustworthiness and Causality through Four Modalities
  di: Lu, Chaochao, et al.
  Pubblicazione: (2024)
• Navigate Beyond Shortcuts: Debiased Learning through the Lens of Neural Collapse
  di: Wang, Yining, et al.
  Pubblicazione: (2024)
• Reinforcing Video Reasoning with Focused Thinking
  di: Dang, Jisheng, et al.
  Pubblicazione: (2025)