arXiv Machine Learning
Coding-Agent Benchmarks Should Match Their Users' Task Flows
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
Hugging Face Trending Papers
6d ago
Coding-Agent Benchmarks Should Match Their Users' Task Flows
The evaluation of coding agents generally strives to be as realistic as possible. In our study, we collect 4,782 agent sessions of real software engineers in JetBrains IDEs, which we call Production S...
arXiv Machine Learning
Jun 30
SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions
arXiv:2606. 30573v1 Announce Type: new Abstract: We introduce SWE-Interact, a new testbed for evaluating coding agents on multi-turn, interactive, user-driven software engineering tasks.
arXiv AI
Jun 9
SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?
arXiv:2606. 07682v1 Announce Type: cross Abstract: AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex environments.
arXiv AI
Jun 30
TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents
arXiv:2606. 28480v1 Announce Type: cross Abstract: As large language models and harness frameworks continue to advance, agents operating in terminals are increasingly capable of performing a broader range of general computer-use tasks beyond coding.
arXiv AI
Jul 14
AgentAbstain: Do LLM Agents Know When Not to Act?
arXiv:2607. 10059v1 Announce Type: new Abstract: Agent systems based on large language models (LLMs) are increasingly deployed for autonomous tasks, yet existing evaluations mostly focus on task success rather than whether agents know when to abstain.
arXiv Machine Learning
Aug 31
RealSWE: A Compositional Evaluation of Coding Agents under Realistic User Requests
arXiv:2608.27831v1 Announce Type: cross Abstract: Coding agents are now commonly evaluated on the SWE-bench family of benchmarks, whose tasks are built from curated GitHub issues--long, structured, a...