Plutus: Pioneering Greek Financial AI in a Global Context
• 8
The Fin AI
Team
community
AI & ML interests
Financial Large Language Models
Recent Activity
View all activity
Papers
Recognition Is Not Grounding: Benchmarking Multilingual Financial Evidence Extraction
Judges Follow the Reference: The Derivation Gap in Verifying Tool-Call Arguments
Articles
Papers by The Fin AI. Each paper's data, models and demos are in its own collection on this org page.
-
AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
Paper • 2606.03031 • Published • 5 -
Herculean: An Agentic Benchmark for Financial Intelligence
Paper • 2605.14355 • Published -
SAHM: A Benchmark for Arabic Financial and Shari'ah-Compliant Reasoning
Paper • 2604.19098 • Published -
FinTrace: Holistic Trajectory-Level Evaluation of LLM Tool Calling for Long-Horizon Financial Tasks
Paper • 2604.10015 • Published
Herculean (arXiv:2605.14355).
FinSeer retriever and StockLLM (arXiv:2502.05878).
EnterpriseArena / CFO-Env (arXiv:2603.23638).
Ebisu (arXiv:2602.01479): JF-ICR and JF-TE.
Conv-FinRe (arXiv:2602.16990).
Reasoning-enhanced financial LLMs, training data and evaluation sets from Fino1 (arXiv:2502.08127).
Datasets introduced or used in MultiFinBen (arXiv:2506.14028).
FinBen (arXiv:2402.12659) and PIXIU/FLARE (arXiv:2306.05443) English tasks, FinMA models and leaderboards.
Spanish tasks and instruction data of FLARE-ES, from 'Dólares or Dollars?' (arXiv:2402.07405).
Public leaderboards maintained by The Fin AI.
-
Open Financial LLM Leaderboard
🏆8Evaluating LLMs on Multilingual Multimodal Financial Tasks
-
Open FinLLM Reasoning Leaderboard
🏆8Evaluating LLMs on complex financial reasoning (Fino1)
-
Open Greek Financial LLM Leaderboard
🏆11Evaluating LLMs on Greek financial tasks
-
IJCAI 2024 FinLLM Leaderboard
🏆4Final results of the IJCAI 2024 FinLLM Challenge
FMDLlama (arXiv:2409.16452) data and the FMD 2025 challenge.
Live trading arena for LLM agents (arXiv:2510.11695).
FinCriticalED (arXiv:2511.14998).
Credit scoring, fraud detection, financial distress and claim analysis tasks from CALM (arXiv:2310.00566).
FinAuditing (arXiv:2510.08886): FinSM, FinRE, FinMR and their subsets.
FinTagging (arXiv:2505.20650): FinNI and FinCL evaluation sets, v1 and v2.
Greek financial benchmark, model and leaderboard from Plutus (arXiv:2502.18772).
Chinese tasks of ICE-PIXIU (arXiv:2403.06249).
Open multimodal financial LLMs from Open-FinLLMs (arXiv:2408.11878).
Papers by The Fin AI. Each paper's data, models and demos are in its own collection on this org page.
-
AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
Paper • 2606.03031 • Published • 5 -
Herculean: An Agentic Benchmark for Financial Intelligence
Paper • 2605.14355 • Published -
SAHM: A Benchmark for Arabic Financial and Shari'ah-Compliant Reasoning
Paper • 2604.19098 • Published -
FinTrace: Holistic Trajectory-Level Evaluation of LLM Tool Calling for Long-Horizon Financial Tasks
Paper • 2604.10015 • Published
Public leaderboards maintained by The Fin AI.
-
Open Financial LLM Leaderboard
🏆8Evaluating LLMs on Multilingual Multimodal Financial Tasks
-
Open FinLLM Reasoning Leaderboard
🏆8Evaluating LLMs on complex financial reasoning (Fino1)
-
Open Greek Financial LLM Leaderboard
🏆11Evaluating LLMs on Greek financial tasks
-
IJCAI 2024 FinLLM Leaderboard
🏆4Final results of the IJCAI 2024 FinLLM Challenge
Herculean (arXiv:2605.14355).
FMDLlama (arXiv:2409.16452) data and the FMD 2025 challenge.
FinSeer retriever and StockLLM (arXiv:2502.05878).
Live trading arena for LLM agents (arXiv:2510.11695).
EnterpriseArena / CFO-Env (arXiv:2603.23638).
FinCriticalED (arXiv:2511.14998).
Ebisu (arXiv:2602.01479): JF-ICR and JF-TE.
Credit scoring, fraud detection, financial distress and claim analysis tasks from CALM (arXiv:2310.00566).
Conv-FinRe (arXiv:2602.16990).
FinAuditing (arXiv:2510.08886): FinSM, FinRE, FinMR and their subsets.
Reasoning-enhanced financial LLMs, training data and evaluation sets from Fino1 (arXiv:2502.08127).
FinTagging (arXiv:2505.20650): FinNI and FinCL evaluation sets, v1 and v2.
Datasets introduced or used in MultiFinBen (arXiv:2506.14028).
Greek financial benchmark, model and leaderboard from Plutus (arXiv:2502.18772).
FinBen (arXiv:2402.12659) and PIXIU/FLARE (arXiv:2306.05443) English tasks, FinMA models and leaderboards.
Chinese tasks of ICE-PIXIU (arXiv:2403.06249).
Spanish tasks and instruction data of FLARE-ES, from 'Dólares or Dollars?' (arXiv:2402.07405).
Open multimodal financial LLMs from Open-FinLLMs (arXiv:2408.11878).