AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Confident and Adaptive Generative Speech Recognition via Conformal Risk Control

Amit Damri, Bracha Laufer-Goldshtein

ICLR 2026 · Large language models › Language, safety and interpretability · approach: Speech and audio language models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LTT adaptive hypothesis selection instead of fixed N=5GER LLaMA-2-7B · TedLium-3average set size: 5 -> 2.48 (-50.08%)Improved
add LTT adaptive selection instead of fixed N=5GER LLaMA-2-7B · TedLium-3WER (%): 7.53 -> 7.52No clear effect
add LTT adaptive selection instead of fixed N=5GER LLaMA-2-7B · CHiME-4WER (%): 6.24 -> 6.37Hurt
add LTT adaptive selection instead of fixed N=5GER LLaMA-2-7B · CommonVoiceWER (%): 8.32 -> 8.51Hurt
add LTT adaptive selection instead of fixed N=5GER LLaMA-2-7B · CommonVoiceaverage set size: 5 -> 3.29 (-34.2%)Improved
raise risk tolerance alpha from 0.21 to 0.24 (ablation)LTT adaptive selection · CommonVoiceWER (%): set size 3.73 -> 2.99; WER 8.53 -> 8.73Mixed
train on constant-5 hypothesis sets instead of dynamic sizes (ablation)GER LLaMA-2-7B · CHiME-4avg WER (%): 8.43 -> 7.79Improved
train on 5 hypotheses instead of 1 hypothesis (ablation)GER LLaMA-2-7B · CHiME-4 (5-hyp test)WER (%): 11.65 -> 6.38Improved
train and test with matched size 3 instead of train-5 (ablation)GER LLaMA-2-7B · CHiME-4 (3-hyp test)WER (%): 6.74 -> 6.58No clear effect
scale GER model from 7B to 13BLTT adaptive selection · TedLium-3WER (%): 7.52 -> 6.46Improved
apply adaptive selection to GPT-3.5-turbo zero-shotLTT adaptive selection · CommonVoiceWER (%): 11.73 -> 11.81No clear effect
apply adaptive selection to speech translation (TER risk)GenTranslate · FLEURS ar->enBLEU: 34.47 -> 33.44Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy