Investor's Last Exam: Can AI Agents Do Equity Investment Without Seeing the Future?
Paper · https://arxiv.org/abs/2610.38342
Financial agents must form forecasts and investment judgments from decision-time evidence, with predictions evaluated against subsequent outcomes. Existing benchmarks assess financial knowledge, research workflows, or forecasting, but evaluating the full research process requires a strictly frozen information set and professional evaluation criteria. We introduce Investor's Last Exam (ILE), an equity-research benchmark led by experienced buy-side investors and fund managers. Its three components are a versioned financial database spanning 3.43 million document versions and nine source categories, supplemented by structured market data, with validated timestamps governing retrieval and computation; a task taxonomy organized around professional analyst deliverables; and an expert-led framework linking scoring criteria to sources, calculations, and assumptions through financial dependency graphs. Evaluation spans 86 assignments and jointly assesses research quality, forecast accuracy, and simulated investment performance. GPT-6 Astra and Claude Opus 5.5 lead overall at 68.1 and 66.9 out of 100, yet even Astra scores only 52.5 on Forecast & Model. Our results show that even state-of-the-art models fail to translate overall research capability into predictive performance. ILE shifts the evaluation unit from isolated capabilities to the evidence--assumption--judgment chain, allowing researchers to locate where analytical advantages fail to translate into forecasts and investment judgments, and providing a traceable, diagnostic foundation for training and evaluating financial agents.
First seen by Oriel on .
AI summary
Investor's Last Exam (ILE) is an equity-research benchmark that evaluates AI agents using frozen decision-time evidence, finding that strong overall research scores do not translate into strong forecasting.
- ILE combines a versioned financial database of 3.43 million document versions across nine source categories with structured market data and validated timestamps.
- Its professional analyst tasks are scored through expert criteria that connect sources, calculations, and assumptions using financial dependency graphs.
- Across 86 assignments, GPT-6 Astra and Claude Opus 5.5 score 68.1 and 66.9 out of 100 overall.
- GPT-6 Astra scores 52.5 on Forecast & Model, illustrating the gap between research quality and predictive performance.
- The benchmark jointly assesses research quality, forecast accuracy, and simulated investment performance.
#ai agents #equity research #financial benchmarks #forecasting
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Points
Score reported by the source.
- Hugging Face Papers: 15, latest observation ; +6 over observed period since
Comments
Comment count reported by the source.
- Hugging Face Papers: 4, latest observation ; +1 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| Hugging Face Papers | 2610.38342 | https://arxiv.org/abs/2610.38342 Read discussion |