AI Paper Atlas

Research map / Large language models / Agents, code and math

FATE: A Formal Benchmark Series for Frontier Algebra of Multiple Difficulty Levels

Jiedong Jiang, Wanyi He, Wang Yuefeng, Guoxiong Gao, Yongle Hu, Jingting Wang, Nailin Guan, Peihao Wu, Bryan Dai, Liang Xiao, Bin Dong

ICLR 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 24 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
Compare natural-language proof stage vs formal Lean stageDeepSeek-R1 · FATE-Haccuracy: 71.0% -> 0.0%Improved
Compare natural-language proof stage vs formal Lean stageDeepSeek-Prover-V2 · FATE-Xaccuracy: 9.0% -> 0.0%Hurt
Move from FATE-M to FATE-H to FATE-XDeepSeek-R1 · FATE-M/H/XNL accuracy: 94.7% -> 71.0% -> 33.0%Hurt
Use general reasoning model instead of specialized proverDeepSeek-R1 vs DeepSeek-Prover-V2 · FATE-HNL pass@1 accuracy: 39.0% -> 71.0%Improved
Remove Lean generation requirement from prompt (ablation)DeepSeek-R1 · FATE-XNL accuracy: 33% -> 39%Improved
Remove Lean requirement and formal statement (ablation)DeepSeek-R1 · FATE-XNL accuracy: 33% -> 43%Improved
Show formal statement in prompt (ablation)DeepSeek-R1 · FATE-XNL accuracy: 43% -> 39%Hurt
Add math-before-Lean output instruction (ablation)DeepSeek-R1 · FATE-XNL accuracy: 33% -> 32%No clear effect
Compare Mathlib hallucination rate across modelsDeepSeek-R1 vs DeepSeek-Prover-V2 · FATE-Herror rate: 35/39 vs 70/71Mixed
Compare misalignment errorsDeepSeek-Prover-V2 vs DeepSeek-R1 · FATE-Hmisalignment: 3/39 vs 0/71Hurt
Increase problem difficulty FATE-M to FATE-XDeepSeek-R1 · FATEoutput tokens: 3845.0 -> 10625.1 -> 16597.4Mixed

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