Coverage-Based Calibration for Post-Training Quantization via Weighted Set Cover over Outlier Channels

Documenti analoghi

• Continual Calibration: Coverage Can Collapse Before Accuracy in Lifelong LLM Fine-Tuning
  di: Shihab, Ibne Farabi, et al.
  Pubblicazione: (2026)
• Calibrated Adaptation: Bayesian Stiefel Manifold Priors for Reliable Parameter-Efficient Fine-Tuning
  di: Shihab, Ibne Farabi, et al.
  Pubblicazione: (2026)
• Selective Risk Certification for LLM Outputs via Information-Lift Statistics: PAC-Bayes, Robustness, and Skeleton Design
  di: Akter, Sanjeda, et al.
  Pubblicazione: (2025)
• Auditing Near-Optimal Policies Can Be Exponentially Hard: Conditional Query Lower Bounds via Occupancy Rashomon Capacity
  di: Shihab, Ibne Farabi, et al.
  Pubblicazione: (2026)
• Detecting Proxy Gaming in RL and LLM Alignment via Evaluator Stress Tests
  di: Shihab, Ibne Farabi, et al.
  Pubblicazione: (2025)