AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

LANGTRAJ: Diffusion Model and Dataset for Language-Conditioned Trajectory Simulation

Wei-Jer Chang, Wei Zhan, Masayoshi Tomizuka, Manmohan Chandraker, Francesco Pittaluga

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Diffusion planning and policies · 11 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 direct language conditioningLangTraj · InterDrive testmADE: 2.65 -> 2.29Improved
add direct language conditioningLangTraj · ProSim-Instruct-520k testmADE: 2.89 -> 2.52Improved
use direct conditioning instead of LLM-based guidanceLangTraj · InterDrive testmADE: 2.70 -> 2.29Improved
use direct conditioning instead of LLM-based guidanceLangTraj · InterDrive testMeta realism: 0.70 -> 0.72Improved
add direct text conditioning to collision guidance (vs LLM guidance)LangTraj · Safety-critical InterDriveCollision rate: 0.33 -> 0.43Improved
add direct text conditioning to collision guidanceLangTraj · Safety-critical InterDriveMap realism: 0.70 -> 0.74Improved
add collision guidanceLangTraj · Safety-critical InterDriveCollision rate: 0.04 -> 0.41Mixed
reduce denoising steps from K=100 to K=5 (ablation)LangTraj · WOSAC validationMeta realism: 0.68 -> 0.68No clear effect
add closed-loop training without teacher forcing (ablation)LangTraj K=5 · WOSAC validationKinematic realism: 0.41 -> 0.38Hurt
add teacher forcing to closed-loop training (ablation)LangTraj · WOSAC validationMap realism: 0.72 -> 0.79Improved
add closed-loop training with teacher forcing (ablation)LangTraj · WOSAC validationMeta realism: 0.68 -> 0.70Improved
raise CFG weight from 0.0 to 2.0 (ablation)LangTraj · CFG weight sweepMap realism: 0.81 -> 0.73Hurt

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