AI Paper Atlas

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

VALLR: Visual ASR Language Model for Lip Reading

Marshall Thomas, Edward Fish, Richard Bowden

ICCV 2025 · Large language models › Language, safety and interpretability · approach: Sign and cross-lingual translation · 17 citations (OpenAlex)

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 phoneme intermediate stage instead of end-to-end sentence prediction (ablation)VALLR · LRS3 testWER (%): 25.6 -> 18.7Improved
use Llama 3.2-3B instead of Llama 3.2-1B for phoneme-to-word LLM (ablation)VALLR · LRS3 testWER (%): 22.8 -> 18.7Improved
fine-tune Llama 3.2-1B on phoneme-word pairs vs GPT-2 Small (ablation)VALLR · LRS3 testWER (%): 33.9 -> 22.8Improved
initialise and freeze the CTC head with Wav2Vec2 ASR weights (ablation)VALLR · LRS3 testWER (%): +0.4% WERHurt
increase visual-encoder training data beyond 30 hours (ablation)VALLR Video-to-Phoneme network · LRS3 testWER (%)No clear effect
add more text training data for the phoneme-to-sentence LLM (ablation)VALLR LLM · LRS3 testWER (%): 18.7 -> 17.5Improved
use VALLR with 30 labelled hours versus self-supervised methods with far more dataVALLR · LRS3 testWER (%): 21.5 -> 18.7 vs USRImproved
apply VALLR on LRS2 versus best self-supervised baselinesVALLR · LRS2 testWER (%): 20.8 vs USR 15.4, RAVEn 17.9Mixed

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