When In-Distribution Gains Fail: Evaluating Weak-to-Strong Reward Models under Preference Shift

Documenti analoghi

• Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models
  di: Cao, Tri, et al.
  Pubblicazione: (2026)
• Vision-and-Language Pretraining
  di: Nguyen, Thong, et al.
  Pubblicazione: (2022)
• READ: Recurrent Adapter with Partial Video-Language Alignment for Parameter-Efficient Transfer Learning in Low-Resource Video-Language Modeling
  di: Nguyen, Thong, et al.
  Pubblicazione: (2023)
• DemaFormer: Damped Exponential Moving Average Transformer with Energy-Based Modeling for Temporal Language Grounding
  di: Nguyen, Thong, et al.
  Pubblicazione: (2023)
• Topic Modeling as Multi-Objective Contrastive Optimization
  di: Nguyen, Thong, et al.
  Pubblicazione: (2024)