AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

FormAct: Agentic Source Editing for Rich-Format Document Generation

Eugene J. Yu, Xingxing Zhang, Yuan Xia, Tao Ge, Xun Wang, FNU Kartik, Vishwas Suryanarayanan, Cheng Yang, Amanda Jiang, Jiayu Ding, Xiangyu Wong, Tengchao Lv and 4 more

ICML 2026 · Multi-agent systems and games › LLM agents · approach: Software engineering agent benchmarks

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Presents FormAct, an agentic system that generates professional rich-format documents from scratch by iteratively refining an HTML source with an editing agent that calls a suite of tools.

Problem

Rich-format documents are costly to author manually, motivating automated generation of professional documents.

Approach

Operates on an HTML source representation and performs iterative source refinement using an editing agent that invokes a suite of tools, rather than generating documents in one shot.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add review-guided refinement loop (ablation)FormAct · RichDocBenchrendering correctness: 4.59 -> 4.81Improved
apply refinement loop to Direct Generation drafts (ablation)DG · RichDocBenchrendering correctness: 4.34 -> 4.82 (+11.1%)Improved
apply refinement loop to Codex drafts (ablation)Codex · RichDocBenchrendering correctness: 4.40 -> 4.92 (+11.8%)Improved
apply refinement loop to TAG drafts (ablation)TAG · RichDocBenchlayout appropriateness: 3.78 -> 3.76 (-0.5%)No clear effect
apply refinement loop to Codex drafts (ablation)Codex · RichDocBenchlayout appropriateness: 3.84 -> 3.98 (+3.6%)Improved
add retrieved templates as in-context examples (TAG vs DG)Direct Generation · RichDocBenchlayout appropriateness: 3.76 -> 3.70Hurt
add retrieved templates as in-context examples (TAG vs DG)Direct Generation · RichDocBenchprofessionalism: 3.75 -> 3.80Improved
use general-purpose coding agent (Codex) instead of direct generationDirect Generation · RichDocBenchrendering correctness: 4.36 -> 4.39No clear effect
construct documents via python-docx library instead of editing HTML sourceFormAct · RichDocBenchlayout appropriateness: 3.92 -> 3.39Hurt
FormAct source editing with template retrieval vs DGDirect Generation · RichDocBenchlayout appropriateness: 3.76 -> 3.92Improved
FormAct compared with Codex in human rankingCodex · RichDocBench human preference (50 examples)rank-1 fraction: 0.140 -> 0.760Improved

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