SpIDER paper boosts code retrieval for coding agents via semantic search plus code graphs
mangahomanga · x · 2026-10-11
SpIDER, an EMNLP 2026 main conference paper, tackles the 'where to look' problem for coding agents in large codebases. It combines global semantic-similarity search with local exploration along the repository's code graph, layering on top of whatever retriever you already use. The authors argue that agents can only fix code they can find, making retrieval half the battle in repo-scale tasks.
- Global search: semantic similarity over the codebase
- Local refinement: following the repo's code graph
- Drop-in: works with existing retrievers
Paper: linked in the original tweet
Related event: SpIDER Paper Helps Coding Agents Locate Code Before Editing(2 posts)→
More from coding & agent
- Self-organizing agent teams hit 66.7% on math benchmarks vs 48.8% for their strongest member — zainhas · 2026-10-12
- Practitioner's verdict on autoresearch: great at speeding experiments, not at frontier runs — iaindunning · 2026-10-12
- levelsio cancels nearly all SaaS subscriptions: money now goes only to AI inference, content, data centers, power and taxes — karlwaldman · 2026-10-12
- Grok Bot is winning the agentic personal assistant race, says dev — Arindam_1729 · 2026-10-12
- Does Codex refuse or cut back work when its own budget estimate runs high? — r618NecessaryStation · 2026-10-12
- Excalidraw for slides is the GOAT: MCP integration plus animations, no AI slop — HamelHusain · 2026-10-12