Epoch confirms GPT5.4 Pro solved a frontier math open problem
TL;DR Highlight
GPT-5.4 Pro is the first to solve a FrontierMath open problem (Ramsey-style hypergraph) — Opus 4.6 and Gemini 3.1 Pro also confirmed it afterward
Who Should Read
Researchers tracking AI capability progress; ML engineers interested in math AI benchmarks
Core Mechanics
- GPT-5.4 Pro solved an Epoch AI FrontierMath open problem confirmed by the problem contributor — to be published
- After a general scaffold was deployed, Opus 4.6 (max), Gemini 3.1 Pro, and GPT-5.4 (xhigh) also solved the same problem
- HN debate: counterarguments to "LLMs cannot produce novel ideas" remain ongoing
Evidence
- Epoch AI official confirmation: problem contributor Prof. Will Brian (UNC Charlotte) validated the solution, publication planned
- Opus 4.6 consumed ~250K tokens — token consumption potentially proxies problem difficulty
How to Apply
- Reference FrontierMath Open Problems (epoch.ai/frontiermath/open-problems) when benchmarking AI model math capabilities
- Use unsolved problem challenges rather than standard benchmarks to assess real capability limits of frontier models
Terminology
Related Papers
Can LLMs model real-world systems in TLA+?
LLM이 TLA+ 명세를 작성할 때 문법은 잘 통과하지만 실제 시스템과의 동작 일치도(conformance)는 46% 수준에 그친다는 걸 체계적으로 검증한 벤치마크 연구로, AI 기반 형식 검증의 현실적 한계를 보여준다.
Natural Language Autoencoders: Turning Claude's Thoughts into Text
Anthropic이 LLM 내부의 숫자 벡터(활성화값)를 직접 읽을 수 있는 자연어로 변환하는 NLA 기법을 공개했다. AI가 실제로 무슨 생각을 하는지 해석하는 interpretability 연구의 새로운 진전이다.
ProgramBench: Can language models rebuild programs from scratch?
LLM이 FFmpeg, SQLite, PHP 인터프리터 같은 실제 소프트웨어를 문서만 보고 처음부터 재구현할 수 있는지 측정하는 새 벤치마크로, 최고 모델도 전체 태스크의 3%만 95% 이상 통과하는 수준에 그쳤다.
MOSAIC-Bench: Measuring Compositional Vulnerability Induction in Coding Agents
티켓 3장으로 쪼개면 Claude/GPT도 보안 취약점 코드를 53~86% 확률로 그냥 짜준다.
Refusal in Language Models Is Mediated by a Single Direction
Open-source chat models encode safety as a single vector direction, and removing it disables safety fine-tuning.
Show HN: A new benchmark for testing LLMs for deterministic outputs
Structured Output Benchmark assesses LLM JSON handling across seven metrics, revealing performance beyond schema compliance.