AI Paper Atlas

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

DNR Bench: Benchmarking Over-Reasoning in Reasoning LLMs

Oluwanifemi Bamgbose, Masoud Hashemi, Sathwik Tejaswi Madhusudhan, Jishnu Sethumadhavan Nair, Aman Tiwari, Vikas Yadav

AAAI 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 17 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 reasoning models (o3-mini-high, DeepSeek-R1) instead of GPT-4o without instructionsDNR Bench prompts · DNR Bench all categoriesaccuracy: Unanswerable 0.42 (GPT-4o) -> 0.00 (O3-mini, DS-R1)Hurt
add explicit instructions to abstain on unanswerable problems (ablation)reasoning models · DNR Bench Unanswerableaccuracy: Claude-3.7-Thinking 0.12 -> 0.82; GPT-4o 0.42 -> 0.98Improved
add explicit instructions (ablation)GPT-4o and Claude-3.7-Thinking · DNR Bench Indifferentaccuracy: GPT-4o 0.84 -> 0.08; Claude 0.80 -> 0.04Hurt
add explicit instructions (ablation)DeepSeek-R1 · DNR Bench Imaginary Referenceaccuracy: 0.00 -> 1.00Improved
add explicit instructions (ablation)O3-mini-high · DNR Bench Unanswerableaccuracy: 0.00 -> 0.02No clear effect
add explicit instructions (ablation)reasoning models · DNR Benchtoken count: up to 70x tokens vs GPT-4o -> 2x to 20xImproved
allow longer reasoning (more tokens)reasoning models · DNR BenchaccuracyHurt
use controlled reasoning budget (GPT-OSS low/medium/high) (ablation)GPT-OSS-120B · DNR Bench Unanswerableaccuracy: low 0.14, medium 0.16, high 0.12No clear effect
use Gemini-2.5-Pro adaptive reasoningGemini-2.5-Pro · DNR Benchtokens and accuracy: over 2000 tokens avg, ~10x GPT-4oHurt

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