{"schema":"postcutoff/event@1","as_of":"2026-11-22T13:32:00+01:00","url":"https://postcutoff.com/e/2025-09-22-alphaevolve-hardness-of-approximation/","md":"https://postcutoff.com/e/2025-09-22-alphaevolve-hardness-of-approximation/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2025-09-22-alphaevolve-hardness-of-approximation","date":"2025-09-22","date_precision":"day","short_title":"AlphaEvolve finds gadgets that prove new NP-hardness of approximation bounds for MAX-k-CUT","deck":null,"takeaway":"Google researchers used AlphaEvolve to discover gadget reductions proving it is NP-hard to approximate MAX-4-CUT within 0.987 and MAX-3-CUT within 0.9649.","category":"science","category_label":"Science & math","importance":2,"confidence":"medium","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":[{"n":1,"title":"Reinforced Generation of Combinatorial Structures (arXiv 2509.18057)","url":"https://arxiv.org/abs/2509.18057","type":"paper","group":"primary","domain":"arxiv.org"},{"n":2,"title":"Google Research: AI as a research partner — advancing theoretical CS with AlphaEvolve","url":"https://research.google/blog/ai-as-a-research-partner-advancing-theoretical-computer-science-with-alphaevolve/","type":"official","group":"primary","domain":"research.google"}],"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Google Research","Google DeepMind"],"title":"AlphaEvolve finds gadgets that prove new NP-hardness of approximation bounds for MAX-k-CUT","summary":"Google researchers used AlphaEvolve to discover gadget reductions proving it is NP-hard to approximate MAX-4-CUT within 0.987 and MAX-3-CUT within 0.9649. They also built near-extremal Ramanujan graphs of up to 163 nodes for average-case hardness results; checking the gadgets was sped up ~10,000×.","key_facts":["arXiv 2509.18057 'Reinforced Generation of Combinatorial Structures'","MAX-4-CUT inapproximability 0.987; MAX-3-CUT 0.9649","Correctness of the final theorems checked by standard (non-AI) verification"],"key_numbers":[],"tags":["theoretical-cs","complexity","alphaevolve"],"science":{"field":"computer-science","subfield":"complexity theory / hardness of approximation","problem":"Inapproximability thresholds for MAX-k-CUT","result":"New NP-hardness of approximation bounds from AI-discovered gadget reductions.","open_since":"","ai_system":["AlphaEvolve"],"human_role":"Human-led with AI tools: researchers framed the gadget search and proved the theorems","verification":{"key":"preprint","label":"Preprint only","text":"Preprint; gadgets verified by exhaustive computation"},"status":"confirmed","shock":""},"body_md":"## What happened\n\nAlphaEvolve searched for finite combinatorial gadgets whose properties imply hardness theorems. Standard verification then turned the found objects into proofs.\n\n## Why it matters\n\nAI-found objects became ingredients of rigorous complexity-theory theorems, not just numeric improvements.","disputed":[],"related":[{"id":"2025-05-14-alphaevolve","url":"https://postcutoff.com/e/2025-05-14-alphaevolve/","date":"2025-05-14","date_precision":"day","short_title":"AlphaEvolve: Gemini-powered agent discovers new algorithms","deck":null,"takeaway":"A concrete example of LLM-based systems making novel discoveries and improving the infrastructure that trains them — an early form of recursive improvement.","category":"science","category_label":"Science & math","importance":4,"confidence":"high","status":{"key":"confirmed","labels":["Result confirmed"]},"sources":4,"official":4,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Google DeepMind"]}],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-09-29","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-29","run":null,"sources_read":null,"updated":"2026-09-29","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":null,"in_training_data":true},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":null,"in_training_data":true},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":null,"in_training_data":true},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":null,"in_training_data":true}],"short_url":null}