AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

LiveNewsBench: Evaluating Web Search Agents with Freshly Curated News

Yunfan Zhang, Kathleen McKeown, Smaranda Muresan

ICML 2026 · Large language models › Retrieval and knowledge · approach: Explanatory and multiple-choice QA benchmarks

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
add web search access (ablation)DeepSeek V3.2 Thinking · LiveNewsBench human-verifiedaccuracy (%): 10.0 -> 84.5Improved
add web search access (ablation)GPT-OSS-120B · LiveNewsBench human-verifiedaccuracy (%): 10.0 -> 27.0Improved
add web search access (ablation)Gemini 3 Pro · LiveNewsBench human-verifiedaccuracy (%): 20.5 -> 60.5Improved
raise search budget from 1 to 7 (ablation)DeepSeek V3.2 (No Thinking) · LiveNewsBench human-verifiedaccuracy (%): 20.0 -> 84.5Improved
raise search budget from 1 to 7 (ablation)GPT-4.1 · LiveNewsBench human-verifiedaccuracy (%): 62.5 -> 66.5Improved
raise search budget from 1 to 7 (ablation)Claude Sonnet 4.5 · LiveNewsBench human-verifiedaccuracy (%): 53.5 -> 67.0Improved
raise search budget from 5 to 7 (ablation)Claude Sonnet 4.5 · LiveNewsBench human-verifiedaccuracy (%): 82.0 -> 67.0Hurt
raise search budget from 5 to 7 (ablation)Gemini 3 Pro · LiveNewsBench human-verifiedaccuracy (%): 60.5 -> 74.5Improved
extend tool budget to 15 searches and 15 visits (ablation)DeepSeek V3.2 Thinking · LiveNewsBench human-verifiedaccuracy (%): 84.5 -> 84.0No clear effect
extend tool budget to 15 searches and 15 visits (ablation)Claude Sonnet 4.5 · LiveNewsBench human-verifiedaccuracy (%): 82.0 -> 81.0No clear effect
use official web search API instead of local agent frameworkGPT-5.2 · LiveNewsBench human-verifiedaccuracy (%): 74.0 -> 90.0Improved
use official web search API instead of local agent frameworkClaude Sonnet 4.5 · LiveNewsBench human-verifiedaccuracy (%): 82.0 -> 40.0Hurt

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