AI Paper Atlas

Research map / Video / Video generation

EmbodiedHead: Real-Time Listening and Speaking Avatar for Conversational Agents

Yu Zhang, Kaiyuan Shen, Yang Li

ECCV 2026 · Video › Video generation · approach: Talking face generation · 2 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 LS-state conditioning to the input (ablation)EmbodiedHead · Ablation: listening-speaking moduleMOD: 2.43 -> 2.10Improved
add FiLM modulation of cross-attention on top of LS-state (ablation)EmbodiedHead · Ablation: listening-speaking moduleMOD: 2.10 -> 1.99Improved
add reference and audio embeddings to AdaLN global condition (ablation)EmbodiedHead · Ablation: global conditionLVE: 6.20 -> 5.76Improved
add motion magnitude condition alone (ablation)EmbodiedHead · Ablation: global conditionLVE: 5.97 -> 6.20Hurt
add second-stage image-domain joint fine-tuning with renderer (ablation)EmbodiedHead · EmbodiedHead speaking test setPSNR: 17.53 -> 18.28Improved
add second-stage image-domain fine-tuning (ablation)EmbodiedHead · EmbodiedHead speaking test setLVE: 5.70 -> 5.71No clear effect
increase sampling steps from 4 to 25 (ablation)EmbodiedHead · Ablation: inference stepsLVE: 5.76 -> 6.18Hurt
increase sampling steps from 1 to 25 (ablation)EmbodiedHead · Ablation: inference stepsFDD: 1.76 -> 1.44Improved
use single-stream audio with LS-state instead of dual-stream audioEmbodiedHead · DualTalk test setLVE: 7.94 -> 7.18Improved
use single-stream audio instead of dual-stream audioEmbodiedHead · DualTalk test setBA: 0.42 -> 0.39Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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