Driving · 2026-10-19

LithoWorld: An Endpoint World Model for Lithography Process-Window Recovery

Botong Zhao, Fang Yu, Xinyuan Chen, Yue Lu

arXiv:2610.23408PDF

AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.

TL;DR

LithoWorld is an endpoint world model for SEM-guided lithography process-window recovery. It uses interior representative targets to determine qualified reference appearance and correction supervision, a causal endpoint alignment branch, and a conditional flow policy that outputs a 2D correction as an action chunk. Trained on 173 FEM SEM groups, it achieves 88.5% qualified-region success and 46.0% interior success on SEM-73 test pairs.

Why it matters

Lithography process drift recovery is an inverse problem where multiple parameter cells can yield qualified morphology, making coordinate-matching insufficient. LithoWorld couples endpoint world modeling with qualified interior supervision and reports both qualified-region success and interior success using real SEM data, enabling evaluation that distinguishes “reaching a qualified cell” from “recovering stronger local window support.”

Method

  • Interior representative supervision: selects a qualified interior endpoint from a 3×3 neighborhood rule over measured qualified cells; missing cells remain unknown and do not count as success.
  • Causal endpoint world model: a frozen DINOv2 encoder feeds a causal Qwen2.5 backbone; a predictive branch aligns current-image features to an endpoint representation using stop-gradient reference features.
  • Conditional flow policy: the flow head outputs a 2D correction action chunk (H=8), with inference using four Euler updates; evaluation accepts any verified qualified endpoint (no reranking).

Limitation

Grid axes have no physical dose/focus calibration, and lookup images are not measurements newly acquired at the continuous predictions; deployment also requires a practical source of the qualified reference.

Abstract (from arXiv)

Lithography process drift can move a fabricated pattern outside its process window, producing unqualified scanning electron microscopy (SEM) morphology. We present LithoWorld, an endpoint world model for SEM-guided parameter recovery. Interior representative targets jointly determine qualified reference appearance and correction supervision. A causal predictive branch aligns currentimage features with endpoint representations, while a conditional flow policy generates a two-dimensional correction as an action chunk. The model learns from 173 groups of measured SEMs organized as focus-exposure maps, preserving unmeasured cells as unknown. On 200 test pairs from the 73-group subset, LithoWorld achieves 88.5% qualified-region success and 46.0% interior success, exceeding the strongest comparison method by 2.5 and 6.0 percentage points, respectively. Ablations separate the effects of target selection, endpoint alignment, and output parameterization. Qualitative SEM retrieval illustrates recovery to qualified interiors without requiring exact reference-coordinate matching.

Related papers

Publication record

Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.

First public release: 2026-10-16 · Indexed by Pub-AI: 2026-10-19 UTC

preprint

LithoWorld: An Endpoint World Model for Lithography Process-Window Recovery

Botong Zhao, Fang Yu, Xinyuan Chen, Yue Lu

Preprint posted: 2026-10-16 · Original record ↗

arxiv
2610.23408
Explicit source versions (1)
  • v1 · 2026-10-16
  1. Preprint available

    Source date: 2026-10-16 · First detected: 2026-10-25 UTC · Historical baseline

    Source evidence (1)
Abstract · arxiv

Lithography process drift can move a fabricated pattern outside its process window, producing unqualified scanning electron microscopy (SEM) morphology. We present LithoWorld, an endpoint world model for SEM-guided parameter recovery. Interior representative targets jointly determine qualified reference appearance and correction supervision. A causal predictive branch aligns currentimage features with endpoint representations, while a conditional flow policy generates a two-dimensional correction as an action chunk. The model learns from 173 groups of measured SEMs organized as focus-exposure maps, preserving unmeasured cells as unknown. On 200 test pairs from the 73-group subset, LithoWorld achieves 88.5% qualified-region success and 46.0% interior success, exceeding the strongest comparison method by 2.5 and 6.0 percentage points, respectively. Ablations separate the effects of target selection, endpoint alignment, and output parameterization. Qualitative SEM retrieval illustrates recovery to qualified interiors without requiring exact reference-coordinate matching.