One Head, Many Models: Cross-Attention Routing for Cost-Aware LLM Selection

Documenti analoghi

• Auto-Eval Judge: Towards a General Agentic Framework for Task Completion Evaluation
  di: Bhonsle, Roshita, et al.
  Pubblicazione: (2025)
• Concept Distillation from Strong to Weak Models via Hypotheses-to-Theories Prompting
  di: Boateng, Emmanuel Aboah, et al.
  Pubblicazione: (2024)
• High-Fidelity Synthetic ECG Generation via Mel-Spectrogram Informed Diffusion Training
  di: Huang, Zhuoyi, et al.
  Pubblicazione: (2025)
• Preventing Catastrophic Forgetting: Behavior-Aware Sampling for Safer Language Model Fine-Tuning
  di: Pham, Anh, et al.
  Pubblicazione: (2025)
• Painless Federated Learning: An Interplay of Line-Search and Extrapolation
  di: Geetika, et al.
  Pubblicazione: (2024)