跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 26 天前AI 評分42

EPSBench:評測用於盈利預測的 LLM 智能體

EPSBench: Benchmarking LLM Agents for Earnings Forecasting

AI 導讀

EPSBench 評測 LLM 智能體能否根據財務文件、業績電話會議紀錄及新聞預測季度盈利,並估算損益表項目。研究涵蓋 86 間美國公司、149 個公司季度,並以分析師共識比較六個模型的預測。三個模型的點預測誤差與分析師共識無可檢測差異;六個模型的名義 80% 預測區間只在 37–70% 的情況下涵蓋實際盈利,營業收入估算誤差解釋了模型間大部分表現差異。

正文

Image 1: Hugging Face's logoHugging Face


Papers

arxiv:2610.38293

Copy markdown

EPSBench: Benchmarking LLM Agents for Earnings Forecasting

Published on Oct 28

·Submitted byImage 2 Liqin Yeon Oct 30

·Image 3: GeorgiaTechGeorgia Institute of Technology

- [x] Upvote -

Authors:

Yanbin Yin,

Chuyu Xu,

Liqin Ye,

Jing Peng,

Chao Zhang

Abstract

The study introduces EPSBench to evaluate AI agents in forecasting quarterly earnings by linking financial evidence to income-statement drivers and comparing model predictions to analyst consensus.

Generated by thinkingmachines/Inkling-Small

Quarterly earnings forecasting requires an agent to connect financial evidence to the income-statement line items that produce a company's reported earnings. We introduce EPSBench, a benchmark for studying this ability across 149 firm-quarters from 86 US companies. Agents read a fixed collection of filings, earnings-call transcripts, and news, then estimate company-specific income-statement drivers with cited evidence. A shared accounting calculator converts these estimates into GAAP diluted earnings per share (EPS), providing a common basis for comparing forecast assumptions. We evaluate six models under a common harness, assessing point forecast error relative to date-matched analyst consensus and checking whether their prediction intervals achieve the stated coverage. Three models show no detectable paired difference in point forecast error from consensus, and three have higher error. Across all six models, nominal 80% intervals contain reported EPS in only 37--70% of runs that return an interval, and most misses fall above the upper bound because the forecasts run low. The driver records trace most of the error gap between models to operating-income estimates.

View arXiv pageView PDFGitHubAdd to collection

Community

Image 4

liqinye

Paper submitter3 minutes ago

Excited to share EPSBench: Benchmarking LLM Agents for Earnings Forecasting!

EPSBench evaluates whether LLM agents can forecast quarterly earnings by reasoning over financial filings, earnings calls, and news, and estimating income-statement drivers with supporting evidence.

Across 149 firm-quarters from 86 US companies, we evaluate six models against analyst consensus.

🔍 Key findings:

  • Three models achieve point forecast errors statistically indistinguishable from analyst consensus.
  • All six models are overconfident: nominal 80% prediction intervals cover actual earnings only 37–70% of the time.
  • Errors in operating-income estimates explain most of the performance differences between models.

We hope EPSBench encourages further research into reliable, evidence-grounded financial forecasting agents. Feedback and discussions are welcome!

Reply

Edit Preview

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

Comment ·Sign up or log in to comment

- [x] Upvote -

Get this paper in your agent:

hf papers read 2610.38293

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.38293 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.38293 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.38293 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

System theme

Company

TOSPrivacyAboutCareers

Website

ModelsDatasetsSpacesPricingDocs

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co