Research
Scaling massive monolithic LLMs continues to yield incredible results. But to truly unlock their ceiling, the next frontier is test-time com…
Scaling massive monolithic LLMs continues to yield incredible results. But to truly unlock their ceiling, the next frontier is test-time compute and dynamic orchestration. Nature solves complex proble
Scaling massive monolithic LLMs continues to yield incredible results. But to truly unlock their ceiling, the next frontier is test-time compute and dynamic orchestration. Nature solves complex problems through collaborative ecosystems. In our new #ICLR2026 paper, we evolved a small coordinator. Instead of competing with the monoliths, it orchestrates them. It learns to dynamically assign Thinker, Worker, and Verifier roles to a pool of frontier models—combining their strengths to hit SOTA on LiveCodeBench. This research is part of the engine powering our new product: Sakana Fugu https://sakana.ai/fugu-beta/ 🐡 What if instead of building one giant AI, we evolved a coordinator to orchestrate a diverse team of specialized AIs? 🐟 Excited to share our new paper: “TRINITY: An Evolved LLM Coordinator”, published as a conference paper at #ICLR2026! Paper: https://arxiv.org/abs/2512.04695 In …
Related
- One of my favorite things about Sakana Fugu is the recursive test-time scaling. When allowed to call itself recursively, it reads its own pr…
- SCATR: Simple Calibrated Test-Time Ranking
- Stability and Generalization in Looped Transformers
- When More Thinking Hurts: Overthinking in LLM Test-Time Compute Scaling
- Adaptive Test-Time Compute Allocation with Evolving In-Context Demonstrations
Source: David Ha (X) | 2026-04-26