Research map / Large language models / Agents, code and math
FrontierCS: Evolving Challenges for Evolving Intelligence
Qiuyang Mang, Wenhao Chai, Zhifei Li, Huanzhi Mao, Shang Zhou, Alexander Du, Hanchen Li, Shu Liu, Edwin Chen, Yichuan Wang, Xieting Chu, Zerui Cheng and 47 more
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 changed | Where | Measured | Effect |
|---|---|---|---|
| raise GPT 5 Thinking reasoning effort from medium to high (ablation) | GPT 5 Thinking · FrontierCS algorithmic track | Score: 15.336 to 12.626 | Hurt |
| raise reasoning effort from low to medium (ablation) | GPT 5 Thinking · FrontierCS algorithmic track | Score | Improved |
| add instruction to keep rectangle state in a 2D array and convert only at the end (ablation) | GPT 5 Thinking · Polyomino Packing | score / zero-score rate: zero-score rate about 30% -> about 10%; scores 20 -- 70 -> 80 -- 85 | Improved |
| sample five attempts instead of one (Score@5 vs Score@1) (ablation) | all 8 frontier models · FrontierCS algorithmic track | Score: +6.40 -- 22.69 points | Improved |
| sample five attempts instead of one (Score@5 vs Score@1) (ablation) | all 8 frontier models · FrontierCS research track | Score: +6.72 -- 21.04 points | Improved |
Closest papers by meaning
- FrontierSmith: Synthesizing Open-Ended Coding Problems at Scale (NeurIPS 2026 · similarity 0.87)
- The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity (NeurIPS 2025 · similarity 0.86)
- CS-Bench: A Comprehensive Benchmark for Large Language Models towards Computer Science Mastery (ICLR 2025 · similarity 0.86)
- LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? (NeurIPS 2025 · similarity 0.85)
- LongCoT: Benchmarking Long-Horizon Chain-of-Thought Reasoning (ICML 2026 · similarity 0.85)
- MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis (NeurIPS 2026 · similarity 0.85)
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.
