Really glad to have been a part of this super cool project... LLMs can verbalize more than just a single confidence number, and we can evaluate their ability to do so!

Many treat uncertainty = a number. At Apple, we're rethinking this: LLMs should output strings that reveal all information of their internal distributions. We find that Reasoning, SFT, CoT can't do it - yet. To get there, we introduce the SelfReflect benchmark. arxiv.org/pdf/2505.20295

Natural idea. Looks like a nice paper too. arxiv.org/abs/2508.21184

Now that @interspeech.bsky.social registration is open, time for some shameless promo! Sign-up and join our Interspeech tutorial: Speech Technology Meets Early Language Acquisition: How Interdisciplinary Efforts Benefit Both Fields. 🗣️👶 www.interspeech2025.org/tutorials ⬇️ (1/2)