Signed in as:
filler@godaddy.com
Github
We believe in open research and community-driven innovation. Our GitHub is home to a growing collect
🌟 Awesome AI Agents
A directory of practical and usable AI agents resources from applications and platforms to frameworks and utilities and other parts of the growing ecosystem
Publications
Reasoning Beyond the Obvious: Evaluating Divergent and Convergent Thinking in LLMs
Most reasoning benchmarks for LLMs emphasize factual accuracy or step-by-step logic. In finance, however, professionals must not only converge on optimal decisions but also generate creative, plausible futures under uncertainty. We introduce ConDiFi, a benchmark that jointly evaluates divergent and convergent thinking in LLMs for financial tasks.
ConDiFi features 607 macro-financial prompts for divergent reasoning and 990 multi-hop adversarial MCQs for convergent reasoning. Using this benchmark, we evaluated 14 leading models and uncovered striking differences. Despite high fluency, GPT-4o underperforms on Novelty and Actionability. In contrast, models like DeepSeek-R1 and Cohere Command R+ rank among the top for generating actionable, insights suitable for investment decisions. ConDiFi provides a new perspective to assess reasoning capabilities essential to safe and strategic deployment of LLMs in finance.
Presented at KDD2025: Workshop on Evaluation and Trustworthiness of Agentic and Generative AI Models, Oral Track
(Joint work with the Government Technology Agency of Singapore)