Skip to content

IGT @ FinMMEval 2026 Task 2: Question-Type Prompting with Targeted Extraction for Multilingual Financial QA

Sep 2026 · Conference and Labs of the Evaluation Forum · pp. 2607-2623 · 1 citation · 13 references
Computer Science

TL;DR

The IGT system for PolyFiQA Task 2 of the FinMMEval Lab at CLEF 2026, a multilingual financial question answering task over English SEC filings and multilingual news articles for four companies, achieves development ROUGE-1 approximately 0.395, a 60% relative improvement over a generic RAG baseline.

Abstract

We present the IGT system for PolyFiQA Task 2 of the FinMMEval Lab at CLEF 2026, a multilingual financial question answering task over English SEC filings and multilingual news articles (English, Chinese, Japanese, Spanish, Greek) for four companies. Our central observation is that the 344 development questions divide into two families requiring fundamentally different approaches: structured numeric types (R&D ratio, cash flow, capital expenditure) are best answered by direct keyword extraction on filing text, while synthesis types (investment strategy, capital allocation, top-three revenue focuses) require rule-based multilingual news passage selection. A dataset analysis reveals that 17-18 of 19 ground-truth reference answers per synthesis type share an exact evidence label prefix, whose unigram tokens contribute directly to ROUGE-1 overlap. The final system achieves development ROUGE-1 approximately 0.395, a 60% relative improvement over a generic RAG baseline (approximately 0.247), and ranks 3rd of 12 teams on the official test set with ROUGE-1 = 0.3071, Precision = 0.2821, and Recall = 0.4044.

View source

Similar papers

2026

TextSentinels @ FinMMEval 2026 Task 2: Chronological Stabilization of a BM25-Grounded Multilingual Financial Question Answering Pipeline

This paper documents the chronological engineering evolution of the chronological engineering evolution of the ROUGE system, beginning with a naïve prompting baseline and progressing through multiple iterations involving retrieval redesign, prompt refinement, multilingual evidence preservation, and infrastructure optim...

D. Thenmozhi, Adithya Sivakumar, Aakash Balasubramanian et al. · 1 citation

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.