EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions
EvoCode-Bench is a newly introduced benchmark designed to evaluate coding agents in multi-turn iterative interactions. It consists of 26 stateful coding tasks and assesses the agents' ability to maintain their codebase as requirements evolve. The study reveals that while some agents perform well in single-round evaluations, they struggle significantly in multi-turn scenarios, highlighting the challenges in specification tracking and regression failures.
- ▪EvoCode-Bench evaluates coding agents through 26 stateful tasks and 227 rounds.
- ▪The benchmark assesses agents' performance over multiple rounds, tracking their ability to adapt to changing requirements.
- ▪Results indicate that most agents achieve only about 50% success in multi-turn metrics, with performance declining over rounds.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.24110 |
| Publication time | Tue, 26 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-26T04:07:43.013Z |
| Last seen | 2026-05-26T04:07:43.013Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | uGsyXOuu4lBU |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.24110 (cs) [Submitted on 22 May 2026] Title:EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions Authors:Haiyang Shen, Xuanzhong Chen, Wendong Xu, Yun Ma, Liang Chen, Kuan Li View a PDF of the paper titled EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions, by Haiyang Shen and 5 other authors View PDF HTML (experimental) Abstract:Coding agents are increasingly used as iterative development partners, but most benchmarks still evaluate one specification followed by one final assessment. This leaves out a basic question: can an agent keep its own codebase working as requirements change? We introduce EvoCode-Bench, a benchmark of 26 stateful coding tasks and 227 evaluated rounds.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
