DAIR.AI 与 MIT 合作提出 SIFT 算法，将代码生成效率提升 10 倍
Recommend read. Love this new arc toward more affordable, more efficient self-improving coding agent...
DAIR.AI 和 MIT 新方法 SIFT，用 LLM 判定器筛选，把代码生成效率提升 10 倍，比 DGM 算法快很多。
DAIR.AI 与 MIT 合作提出的 SIFT 算法，通过 LLM 判定器筛选候选方案，将 CPU 小时数从 224 小时降至 50 小时以下，在 Polyglot 基准上达到 35.1% 的成绩，比 DGM 算法快 10 倍。在 TerminalBench 测试中，使用 gpt-5.4 作为判定器，将 29.2% 的初始模型提升至 36.7%。
Recommend read. Love this new arc toward more affordable, more efficient self-improving coding agent...
Recommend read. Love this new arc toward more affordable, more efficient self-improving coding agents. DAIR.AI @dair_ai Banger paper from MIT and Sakana AI. They show that self-improving coding agents work. The best part is that their approach, Self-Improvement via Fast Tree-search (SIFT), runs at a tenth of the CPU hours of DGM. They reach 35.1 percent on Polyglot with o3-mini after 30 expansions. DGM reaches 30.7 percent after 80 nodes of tree search. SIFT does it in under 50 CPU hours and under 5 hours of wall clock. The Qwen3-30B configuration runs its full search at 224 CPU hours and $34 of API spend, a tenth of the DGM baseline. The saving comes from where the money goes. Benchmark evaluation is the runtime bottleneck, so an LLM judge ranks candidate self-modifications first and only promising candidates get evaluated. Judge quality decides the run. On TerminalBench, gpt-5.4-high as the pairwise judge finds a 36.7 percent agent against a 29.2 percent starting point. gpt-5 finds 34.5 percent, and its top-ranked candidate is not the best agent its search produced. Paper: academy.dair.ai/papers/self-im… 🔗 View Quoted Tweet 💬 3 🔄 0 ❤️ 3 👀 2043 📊 4 ⚡