DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck

Shubham Agarwal, Akshat Gupta, Mike Heddes, Dilara Soylu, Rui Meng, Patrick Tser Jern Kon, Jiefeng Chen, Mert Cemri, Shu Liu, Abe Gonzalez, Akanksha Jain, Bhavana Dalvi Mishra, Linfeng Song, Vilobh Meshram, Xiangru Tang, Jinsung Yoon, Aditya Parameswaran, Ion Stoica, Vahab Mirrokni, Burak Gokturk, Parthasarathy Ranganathan, Tomas Pfister, Nanyun Peng, Chun-Liang LiarXiv 2026 · arXiv

TL;DR

DuoDiscover is a new search system that scales autonomous engineering discovery by overcoming the "evaluation bottleneck"—the massive time cost of running full benchmarks. By having a fleet of agents co-discover fast proxy evaluators (called "yardsticks") alongside optimization targets, it achieves significant speedups in complex systems like vLLM and CPU microarchitecture, often outperforming both human experts and state-of-the-art AI baselines.

What Problem Does This Paper Address?

In the quest for autonomous AI discovery—where agents optimize code, algorithms, or hardware—the primary hurdle isn't just writing the code; it's knowing if the code is actually better.

In real-world engineering, a "true evaluation" (like measuring the throughput of a large-scale serving engine or simulating a CPU design) can take anywhere from 15 minutes to 28 hours. If your total search budget is 20 hours, you can only afford a handful of these evaluations. This makes traditional search methods, which rely on constant feedback, painfully slow and ineffective. The authors identify this as the evaluation bottleneck.

Methods

DuoDiscover tackles this bottleneck through three main mechanisms:

  1. Co-discovery of Targets and Yardsticks: Instead of evaluating every small code change against the full, expensive benchmark, each worker agent identifies a specific target (a function or module) and builds a yardstick—a fast, tailored proxy evaluator (e.g., a microbenchmark or a cost model). The agent then "hill-climbs" the yardstick in a fast inner loop (seconds) before sending promising candidates to the slow outer loop (hours) for true evaluation.
  2. Fidelity Testing: Proxy evaluators are prone to "reward hacking," where an agent finds a way to improve the proxy score without actually improving the true objective. DuoDiscover continually runs fidelity tests using Kendall’s $\tau$ rank correlation to ensure the yardstick's ranking still matches the true objective. If a yardstick drifts, it is invalidated.
  3. Bottom-up Fleet Orchestration: A lightweight orchestrator manages a fleet of independent agents. It doesn't follow a rigid plan; instead, it terminates stalled agents, briefs others on successful discoveries, and "seeds" new agents with confirmed improvements from their peers.
The DuoDiscover loop showing the fast inner yardstick loop and slow outer true evaluation loop

Datasets

The authors evaluated DuoDiscover across a diverse range of benchmarks spanning the entire computing stack:

  • Repository-level Optimization: FormulaCode (scientific Python) and SWE-fficiency (popular libraries like NumPy/pandas).
  • Kernel Suites: SOL-Exec (235 GPU kernels), JAXBench (TPU operators), TPU-Inference, and FA4 (Flash Attention 4).
  • Real-world Systems: The vLLM serving engine, SQLite database, and zstd compression library.
  • Hardware Design: The DPC4 (Data Prefetching Championship) CPU microarchitecture task.

Key Insights and Ideas

  • Yardsticks are Target-Specific: Building a single fast proxy for an entire codebase is nearly impossible. However, building a fast proxy for a single function is much more feasible. DuoDiscover exploits this by letting agents define their own narrow scope.
  • Fidelity is a Moving Target: A yardstick that is accurate at the start of a search might become inaccurate as the code evolves. The "aha moment" here is that validation must be continuous, not just a one-time setup.
  • Search Efficiency over Token Volume: DuoDiscover often uses significantly fewer true evaluations and tokens than baselines like Claude Code to reach the same performance level because it spends its "thinking budget" on high-fidelity candidates.
  • Bottom-up Coordination: By allowing agents to pick their own targets and then having the orchestrator combine their gains, the system avoids the need for a complex central planner that understands the entire system architecture.

Key Results

DuoDiscover consistently outperformed human experts and specialized AI baselines:

  • vLLM: Achieved a 1.92× geometric mean gain across four service-level objectives (latency, power, throughput), while the baseline Claude Code only reached 1.07×.
  • SOL-Exec: Ranked 1st on 121 of 235 kernels on the public leaderboard, significantly ahead of proprietary and human-tuned entries.
  • Chip Design (DPC4): Surpassed the previous SOTA (ArchAgent v2) and the best human design. It delivered a 6.3% performance gain in 16.7 days, compared to ArchAgent v2’s 3.8% gain which took over 40 days.
  • SWE-fficiency: Doubled the average speedup achieved by human expert patches.
vLLM speedup over time, showing DuoDiscover's rapid gains compared to Claude Code

Limitations and Future Work

  • Warmup Overhead: Constructing yardsticks and running initial fidelity tests requires an upfront time investment. Consequently, DuoDiscover is less advantageous for extremely short search budgets where there isn't enough time to amortize the cost of building proxies.
  • Yardstick Availability: The system relies on the agent's ability to create a faithful proxy. If a specific optimization target is so deeply coupled with the rest of the system that no fast yardstick can be built, DuoDiscover falls back to standard, slow search.
  • Future Directions: The authors suggest extending this "co-discovery" approach to even more diverse domains beyond software and chip design, potentially into experimental sciences where physical lab evaluations are the bottleneck.

Summary updated

Abstract

Evaluation has been central to advances in AI, and it is equally critical to autonomous discovery. Yet scaling autonomous discovery to real-world engineering challenges faces a fundamental bottleneck: the cost of evaluating progress. In real settings, a single true evaluation, such as a full benchmark or serving trace, can take hours, so a search can afford only a handful of true evaluations. We present DuoDiscover, a search system that jointly discovers what to optimize and how to evaluate progress efficiently, through three mechanisms. First, each worker in a fleet of coding agents co-discovers an optimization target and a fast, tailored proxy evaluator, which we call a yardstick. The agent then hill-climbs the yardstick rather than the true evaluation. Second, since search can exploit gaps between a yardstick and the true objective, fidelity tests continually recheck each yardstick's ranking and invalidate any that drift. Third, the fleet is bottom-up, with no central decomposition of the problem. A lightweight orchestrator spawns and terminates agents, briefs running ones, and seeds new agents with others' confirmed work. DuoDiscover enables autonomous optimization from software to chip architecture. We demonstrate DuoDiscover on two repository-level benchmarks and four GPU and TPU kernel suites covering hundreds of kernels. On SWE-fficiency, it doubles the average speedup of human experts. On SOL-Exec, it ranks first on 121 of 235 kernels and in the top three on 212. On the heavily optimized vLLM and SQLite engines, DuoDiscover achieves 1.92$\times$ and 1.38$\times$ gains over the base code. On a CPU microarchitecture optimization task with evaluations of up to 28 hours, DuoDiscover surpasses both AlphaEvolve-based ArchAgent v2 and the best human design, with 1.65$\times$ ArchAgent v2's gain in 16.7 days versus more than 40 days, advancing the frontier of autonomous chip design.
BibTeX
@misc{agarwal2026duodiscover,
  author = {Shubham Agarwal and Akshat Gupta and Mike Heddes and Dilara Soylu and Rui Meng and Patrick Tser Jern Kon and Jiefeng Chen and Mert Cemri and Shu Liu and Abe Gonzalez and Akanksha Jain and Bhavana Dalvi Mishra and Linfeng Song and Vilobh Meshram and Xiangru Tang and Jinsung Yoon and Aditya Parameswaran and Ion Stoica and Vahab Mirrokni and Burak Gokturk and Parthasarathy Ranganathan and Tomas Pfister and Nanyun Peng and Chun-Liang Li},
  title = {{DuoDiscover: Scaling Autonomous Discovery Beyond the Evaluation Bottleneck}},
  year = {2026},
  eprint = {2610.17184},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.17184},
}
Figure