The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection

Documenti analoghi

• COME: Test-time adaption by Conservatively Minimizing Entropy
  di: Zhang, Qingyang, et al.
  Pubblicazione: (2024)
• Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
  di: Zhang, Qingyang, et al.
  Pubblicazione: (2025)
• Selective Learning: Towards Robust Calibration with Dynamic Regularization
  di: Han, Zongbo, et al.
  Pubblicazione: (2024)
• Out-Of-Distribution Detection with Diversification (Provably)
  di: Yao, Haiyun, et al.
  Pubblicazione: (2024)
• Semantic Energy: Detecting LLM Hallucination Beyond Entropy
  di: Ma, Huan, et al.
  Pubblicazione: (2025)