Skip to content
Papers.

SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks? research paper by Scale AI, 2025

Scale AI · Sep 21, 2025 · Agents and evaluation · 21 upvotes 1 year ago

Read on arXiv

What it shows

SWE-Bench Pro is a challenging benchmark for coding models, featuring complex, enterprise-level problems that require substantial code modifications, with performance evaluations showing significant limitations in current models.

Hugging Face's summary; not yet checked by hand.

More from Scale AI

All 25
PaperCitations
Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent FailuresContinual Search improves automated root-cause attribution in long agent execution traces by iteratively prompting diagnosis until unresolved evidence is found.Agents and evaluation · Sep 20262 weeks ago-
SteerDuplex: Steerable Duplex Speech Dialogue ModelsA full-duplex speech model that follows spoken instructions on tone, persona, pace and voice, with a benchmark of 390 prompts to test it.Multimodal and robotics · Sep 20262 weeks ago-
Studying Without a Syllabus: Task-Agnostic Environment PreprocessingAn agent can explore unfamiliar environments without task-specific guidance to build reusable artifacts that reduce later inference costs, though larger study budgets do not always improve results.Agents and evaluation · Sep 20262 weeks ago-
HarnessOpt-Bench: Evaluating LLMs at Harness OptimizationHarnessOpt-Bench measures how well frontier LLMs iteratively improve agent harnesses under constrained evaluation budgets, revealing substantial variation across models and tasks.Agents and evaluation · Aug 20267 weeks ago-
Model or Harness? An Interaction-Centric Taxonomy for Localizing Agent FailuresAn interaction-centric taxonomy localizes agent failures to specific component interactions to guide targeted repairs across diverse architectures.Agents and evaluation · Jul 20268 weeks ago-
SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding SessionsSWE-Interact presents a testbed that evaluates coding agents in realistic multi-turn, user-driven software engineering scenarios, revealing significant gaps between single-turn performance and interactive task completion.Agents and evaluation · Jun 20262 months ago-
Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVRPOW3R is a policy-aware framework for reinforcement learning with rubric-based rewards that adapts criterion weights during training to improve policy optimization while preserving human-defined criteria importance.Training and scaling · May 20264 months ago-
Reward Hacking in Rubric-Based Reinforcement LearningResearch examines reward hacking in rubric-based reinforcement learning, identifying verifier failure and rubric-design limitations as key sources of divergence between training and evaluation metrics.Reasoning · May 20264 months ago-

Other agents and evaluation papers

Topic
PaperCitations
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic CapabilitiesGemini 2.X model family, including Gemini 2.5 Pro and Flash, offers superior coding, reasoning, and multimodal understanding capabilities across a range of computational efficiencies.Google · Jul 2025 · Unverified1 year ago4,134
DeepSeek-V3.2: Pushing the Frontier of Open Large Language ModelsDeepSeek-V3.2 introduces DeepSeek Sparse Attention and a scalable reinforcement learning framework, achieving superior reasoning and performance compared to GPT-5 and Gemini-3.0-Pro in complex reasoning tasks.DeepSeek · Dec 2025 · Unverified9 months ago761
WebWatcher: Breaking New Frontier of Vision-Language Deep Research AgentWebWatcher, a multimodal agent with enhanced visual-language reasoning, outperforms existing agents in complex visual and textual information retrieval tasks using synthetic trajectories and reinforcement learning.Alibaba (Qwen) · Aug 2025 · Unverified1 year ago117
SkillOpt: Executive Strategy for Self-Evolving Agent SkillsSkillOpt introduces a systematic text-space optimizer for agent skills that trains skills as external agent state with stable updates and zero deployment inference overhead, achieving superior performance across multiple benchmarks and execution environments.Microsoft · May 2026 · Unverified4 months ago85
Nemotron 3 Nano: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic ReasoningWe present Nemotron 3 Nano 30B-A3B, a Mixture-of-Experts hybrid Mamba-Transformer language model.NVIDIA · Dec 2025 · Unverified9 months ago81
AgentFold: Long-Horizon Web Agents with Proactive Context ManagementAgentFold, a novel proactive context management paradigm, enhances long-horizon task performance through dynamic context folding, achieving superior results on benchmarks compared to larger models and proprietary agents.Alibaba (Qwen) · Oct 2025 · Unverified11 months ago77
About this paper
Authors
Xiang Deng, Jeff Da, Edwin Pan and 16 more
arXiv
2509.16941 · PDF
Citations
Not counted yet · Semantic Scholar
Upvotes
21 · Hugging Face
Code
github.com/scaleapi/SWE-bench_Pro-os
Lab
Scale AI · on Companies · on Acquisitions

Changes

What changed
New paperAdded to the listSep 26, 2026today

New papers by email

Monday afternoons, only in weeks with new papers from the labs.

Double opt-in. Unsubscribe any time.