AI Paper Atlas

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

SWERank: Software Issue Localization with Code Ranking

Revanth Gangi Reddy, Tarun Suresh, JaeHyeok Doo, Ye Liu, Xuan-Phi Nguyen, Yingbo Zhou, Semih Yavuz, Caiming Xiong, Heng Ji, Shafiq Joty

ICLR 2026 · Large language models › Agents, code and math · approach: Software engineering agents

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
apply top-K consistency filtering (K=20) to SweLoc training data instead of no filtering (ablation)SweRankEmbed-Small · SWE-Bench-Litelocalization accuracyImproved
increase training data proportion (even 5%) of filtered SweLoc (ablation)SweRankEmbed-Small · SWE-Bench-Litelocalization accuracyImproved
finetune CodeRankEmbed on SweLocCodeRankEmbed · SWE-Bench-LiteFunc. Acc@10 (%): 59.5 -> 72.3Improved
finetune Arctic-Embed on SweLocArctic-Embed · SWE-Bench-LiteFunc. Acc@10 (%): 53.7 -> 71.9Improved
finetune Qwen-2.5-Text 32B reranker on SweLocQwen-2.5-Text (32B) · SWE-Bench-LiteFunc. Acc@10 (%): 82.5 -> 86.1Improved
finetune Qwen-2.5-Text 7B reranker on SweLocQwen-2.5-Text (7B) · SWE-Bench-LiteFunc. Acc@5 (%): 75.2 -> 75.6No clear effect
add SweRankLLM reranking on top of SweRankEmbed-Large (ablation)SweRankEmbed-Large · SWE-Bench-Lite functionAcc@10 (%): 82.12 -> 88.69Improved
use SweRank localization instead of LocAgent for downstream repairSWE-Fixer · SWE-Bench-LiteRepair Pass@1 (%): 22.6 -> 24.5Improved
add second iteration of hard-negative mining vs first iteration (ablation)SweRankEmbed · SWE-Bench-Litelocalization accuracyImproved
reduce embedding dimension from 1024 to 64 (Matryoshka) (ablation)Qwen3-Embedding-0.6B · SWE-Bench-LiteAcc@10 (%): 75.18 -> 60.22Hurt
scale encoder from SweRankEmbed-Large (7B) to Qwen3-Embedding-8B (deeper) (ablation)retriever · SWE-Bench-LiteAcc@10 (%): 82.12 vs 83.94No clear effect
add SweLoc finetuning to RankZephyr-finetuned Llama-3.1 8B reranker (ablation)Llama-3.1 8B Instruct reranker · SWE-Bench-LiteAcc@10 (%): 80.29 -> 85.77Improved

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