You can't: brief exchange underscores the unsolved problem of reliable LLM confidence
josephdviviano · x · 2026-09-25
@capeandcode notes the enduring struggle of getting reliable confidence scores from LLMs—not random numbers but ones you can trust. @josephdviviano's blunt reply: "You can't."
More from Models
- Apple drops new HF model: Qwen3.5-9B finetune that turns long docs into page images to save tokens — yoobinray · 2026-09-25
- Alexandr Wang amplifies Muse skepticism post calling the model 'underwhelming and useless' — maheen_sohail · 2026-09-25
- The AI Meme: Wait Until Your Project Is 78% Done, Then They Nerf the Model — BLUECOW009 · 2026-09-25
- Open Weights Beat Black-Box APIs on Security: Sandboxing Is an Engineering Problem — ypatil125 · 2026-09-25
- Zyphra open-sources ZUNA1.1 EEG foundation model, advancing noninvasive thought-to-text — burny_tech · 2026-09-25
- Contrastive-LM releases CLM-v0.1-8B, a Qwen3-8B-based reranker trending on Hugging Face — Contrastive-LM · 2026-09-25