AI Paper Atlas

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

Asking What Matters: Reward-Driven Clarification for Software Engineering Tasks

Sanidhya Vijayvargiya, Vijay Viswanathan, Graham Neubig

ICML 2026 · Large language models › Agents, code and math · approach: Software engineering agents · 0 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
add Stage 3 answerability reward (ablation)CLARITI · SWE-Bench Verified underspecified (250 issues)task success rate (%): 24.4 -> 36.8Improved
add Stage 4 task relevance reward (ablation)CLARITI · SWE-Bench Verified underspecified (250 issues)task success rate (%): 32.0 -> 36.8Improved
add RL with staged rewards after SFT (ablation)Qwen3-8B · SWE-Bench Verified underspecified (250 issues)task success rate (%): 24.2 -> 36.8Improved
add clarification questions from CLARITISeed-OSS-36B agent · SWE-Bench Verified underspecified (250 issues)task success rate (%): 22.4 -> 36.8Improved
use CLARITI instead of GPT-5 for clarificationagent pipeline · SWE-Bench Verified underspecified (250 issues)average number of questions: 5.1 -> 3.0Improved
use CLARITI instead of GPT-5agent pipeline · SWE-Bench Verified underspecified (250 issues)task success rate (%): 35.6 -> 36.8No clear effect
use GPT-5 instead of GPT-5 Nano for questionsagent pipeline · SWE-Bench Verified underspecified (250 issues)task success rate (%): 29.6 -> 35.6Improved
remove information categories from issue (underspecify)Seed-OSS-36B agent · SWE-Bench Verified rewrites (700 instances)task success rate (%): 43.8 -> 23.7Hurt
increase number of clarification questionsclarification sets · SWE-Bench Verified underspecified (500 issues)task success rateNo clear effect
add Stage 4 relevance weighting: share of Error Information questionsCLARITI · question category distributionshare of questions (%): 10.2% -> 26.4%Improved

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