Find It Fix It
FIFI 2026

Shared Task on Scientific Title Retrieval & Rewriting โ€” can your system trace an LLM-rewritten title back to its source?


โš™ Technical

"A Transformer-Based Framework for Cross-Domain Semantic Title Alignment"

๐Ÿ“– Accessible

"How AI Learns to Match Research Topics Across Different Fields"

๐Ÿ”ฅ Catchy

"Lost in Translation? AI Finally Bridges the Gap Between Research Silos"


๐Ÿ“ข The results for Task 2 are now available (Scroll Down)

3
Style categories
2
Subtasks
10
Variants per style

๐ŸŽ‰ Task 1 & 2 Results Released

Task 2 Results
Task 2 Task 2 Results
\

What is FIFI?

LLMs can rewrite scientific titles into three styles โ€” Technical, Accessible, and Catchy. FIFI challenges participants to reverse that transformation: given only a reformulated title, retrieve or reconstruct the original research title.

๐Ÿ”

Subtask 1: Find It

Rank up to 10 candidate titles by likelihood of being the true source. Evaluated with MRR@10.

โœ๏ธ

Subtask 2: Fix It

Reconstruct the original title from its stylistic variant. Evaluated with F1 Score & BERTScore.