AWS's EasyProver solves 669/672 Putnam problems, plus new harder benchmark Aegis
amazonwebservices · hf · 2026-10-30
AWS researchers present "Scaling Lean Theorem Proving for Real-World Code Verification" with two systems:
- EasyProver: a Lean proving system using recursive agents with isolated contexts, asynchronous search, parallel tactic probing, and an elastic elaboration service. With no additional training, an off-the-shelf frontier model proves 669/672 Putnam-Bench problems and all six IMO 2026 formalizations.
- Aegis: since existing benchmarks near saturation, it converts competitive programming problems into much harder verification tasks over real implementations. EasyProver resolves 597/600 NTP4VC verification conditions and all 189 VERINA tasks by proof or refutation.
The pipeline writes safe Rust solutions, translates them to Lean with AENAES, and can support RL and real-world code verification research.
More from coding & agent
- Simlock: open-source control plane gives parallel coding agents their own iOS/Android simulators — Scobleizer · 2026-10-30
- Clear Thought MCP Server adds structured reasoning and mental models to AI clients — modelcontextprotocol · 2026-10-30
- NSE Stock Insights MCP brings AI analysis to India's Nifty 500 stocks — modelcontextprotocol · 2026-10-30
- OpenRouter rebuilds provider routing: 46% fewer capacity errors, 16.9% faster TTFT — steipete · 2026-10-30
- Databricks launches Agent Memory, tripling agent pass rates with 31% fewer tokens — matei_zaharia · 2026-10-30
- Build a team of AI agents: Grok for content, Claude Code for marketing, one Obsidian vault as shared brain — EXM7777 · 2026-10-30