AI Paper Atlas

Research map / Large language models / Agents, code and math

On Path to Multimodal Historical Reasoning: HistBench and HistAgent

Jiahao Qiu, Fulian Xiao, Yimin Wang, Yuchen Mao, Yijia Chen, Xinzhe Juan, Siran Wang, Xuan Qi, Tongcheng Zhang, Zixin Yao, Jiacheng Guo, Yifu Lu and 13 more

ICML 2026 · Large language models › Agents, code and math · approach: Agent instruction-following benchmarks · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use HistAgent (history-specific tools and manager-specialist agents) instead of generic Open Deep Research agentGPT-4o agent · HistBenchpass@2 accuracy: 25.12 -> 36.47 pass@2; 19.57 -> 28.50 pass@1Improved
wrap GPT-4o in an agent framework (ODR-smolagents) versus GPT-4o with web searchGPT-4o · HistBenchaccuracy: 18.60 -> 19.57Improved
use HistAgent (GPT-4o) versus stronger reasoning model with searcho4-mini-high with web search · HistBenchpass@1 accuracy: 30.19 vs 28.50Hurt
use HistAgent versus ODR-smolagents on HLE history questionsGPT-4o agent · HLE History subsetpass@1: 17.86 -> 28.57Improved
use HistAgent versus ODR-smolagents on general GAIA tasksagent with strong backbone · GAIA validationpass@1: 55.15 -> 60.00Improved
HistAgent versus ODR-smolagents on GAIA Level 3agent · GAIA validation Level 3pass@1: 34.62 -> 34.62No clear effect
use HistAgent on multimodal questions versus text-onlyHistAgent (GPT-4o) · HistBench Level 1 modality splitpass@1: multimodal 35.63 vs ODR 11.49; text-only 21.52 vs ODR 21.52Mixed
allow second attempt (pass@2 versus pass@1)HistAgent (GPT-4o) · HistBenchaccuracy: 28.50 -> 36.47Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy