ASI-Evolve: AI Accelerates AI

Weixian Xu; Tiantian Mi; Yixiu Liu; Yang Nan; Zhimeng Zhou; Lyumanshan Ye; Lin Zhang; Yu Qiao

cs.AIMar 31, 2026

ASI-Evolve: AI Accelerates AI

Authors: Weixian Xu, Tiantian Mi, Yixiu Liu, Yang Nan, Zhimeng Zhou, Lyumanshan Ye, Lin Zhang, Yu Qiao,

Organizations: SJTU · SII · GAIR

Abstract

Can AI accelerate the development of AI itself? While recent agentic systems have shown strong performance on well-scoped tasks with rapid feedback, it remains unclear whether they can tackle the costly, long-horizon, and weakly supervised research loops that drive real AI progress. We present ASI-Evolve, an agentic framework for AI-for-AI research that closes this loop through a learn-design-experiment-analyze cycle. ASI-Evolve augments standard evolutionary agents with two key components: a cognition base that injects accumulated human priors into each round of exploration, and a dedicated analyzer that distills complex experimental outcomes into reusable insights for future iterations. To our knowledge, ASI-Evolve is the first unified framework to demonstrate AI-driven discovery across three central components of AI development: data, architectures, and learning algorithms. In neural architecture design, it discovered 105 SOTA linear attention architectures, with the best discovered model surpassing DeltaNet by +0.97 points, nearly 3x the gain of recent human-designed improvements. In pretraining data curation, the evolved pipeline improves average benchmark performance by +3.96 points, with gains exceeding 18 points on MMLU. In reinforcement learning algorithm design, discovered algorithms outperform GRPO by up to +12.5 points on AMC32, +11.67 points on AIME24, and +5.04 points on OlympiadBench. We further provide initial evidence that this AI-for-AI paradigm can transfer beyond the AI stack through experiments in mathematics and biomedicine. Together, these results suggest that ASI-Evolve represents a promising step toward enabling AI to accelerate AI across the foundational stages of development, offering early evidence for the feasibility of closed-loop AI research.

Figures & tables

Figure 1: Representative scientific automation settings in the $L_{\text{task}}=\langle C_{\text{exec}},S_{\text{space}},D_{\text{feedback}}\rangle$ space.

Figure 2: Asi-Evolve pipeline: in each round, the system samples context nodes from database, retrieves relevant cognition items via embedding search, generates a new candidate program, runs an experiment-specific evaluation script under timeouts, and summarizes results into an analysis report that is stored back into the database for future rounds.

	Human Discovered			AI Discovered
Benchmarks	DeltaNet	Gated- DeltaNet	Mamba2	PG	C	FG	H	AM
Development
Wiki ppl $\downarrow$	17.00	16.84	16.66	16.18	16.05	15.77	16.65	16.26
LMB ppl $\downarrow$	13.63	13.31	13.33	12.62	13.45	12.34	13.06	13.75
LMB	45.47	46.26	46.24	47.60	46.13	47.53	46.56	45.04
PIQA	73.12	74.10	73.78	72.91	74.37	72.91	74.37	74.10

Table 1: Top block: 10 development benchmarks, used in our exploration stage; middle block: 6 generalization benchmarks for out-of-distribution testing. Bold indicates the best result and underline is the suboptimal one. Model abbreviations are as follows: PG = PathGate-FusionNet, C = Content-SharpRouter, FG = FusionGated-FIRNet, H = Hier-GateNet, and AM = AdaMulti-PathGateNet.

	Human Discovered				AI Discovered
Benchmark	Fineweb-Edu	Ultra-Fineweb	DCLM	Nemotron-CC	$\textbf{Nemotron-CC}_{\textbf{ASI}}$	$\textbf{Nemotron-CC}_{\textbf{ASI+}}$
BBH	3.01	7.42	24.16	26.82	26.69	26.16
ARC-E	73.39	73.96	75.13	74.94	77.55	78.59
ARC-C	43.45	43.77	45.02	43.52	48.41	49.32
MMLU	28.38	25.53	28.54	27.49	32.55	46.13
AGIEval	16.96	17.72	17.90	18.15	18.30	18.21

Table 2: Benchmark comparison of models trained on different pretraining corpora and curation strategies. All models are 3B parameters trained on 500B tokens. Columns are grouped into Human baselines and ASI-Evolve discovered datasets.

Method	Math500	AMC32	AIME25	AIME24	OlympiadBench
GRPO ${}_{\text{Human}}$	82.0	67.5	20.00	20.00	45.92
Algorithm1 ${}_{\text{ASI-Evolve}}$	85.6	80.0	29.58	31.67	50.96
Algorithm2 ${}_{\text{ASI-Evolve}}$	84.6	77.5	23.75	23.33	48.74
Algorithm3 ${}_{\text{ASI-Evolve}}$	84.8	77.5	30.00	30.00	49.18
Algorithm4 ${}_{\text{ASI-Evolve}}$	82.4	75.0	20.00	20.00	46.81
Algorithm5 ${}_{\text{ASI-Evolve}}$	82.0	72.5	23.33	20.00	45.62

Table 3: Performance comparison of evolved RL algorithms and the GRPO baseline on mathematical reasoning benchmarks. All methods were trained on Qwen-3-14B-base within the SIIRL framework using the Skywork OR1 dataset and evaluated after 250 training steps.

Framework	Model	Rounds	Best Score
AlphaEvolve ( Novikov et al., 2025 )	Gemini 2.0 Flash + Claude 3.7	—	2.6359
OpenEvolve ( Sharma, 2025 )	Gemini 2.0 Flash + Claude 3.7	460	2.6343
LoongFlow ( Team, 2024 )	DeepSeek-R1-250528	—	2.6360
SkyDiscover	GPT-5	89	2.6360
ASI-Evolve (Ours)	GPT-5-mini	17	2.6360

Table 4: Circle packing comparison across evolutionary frameworks (26 circles in a

1\times 1

unit square; higher sum of radii is better). “Rounds” denotes the number of evolution steps to reach the reported best score; “—” indicates the value was not reported.

Panel of Figure 3 — Figure 3: Comparison experiments. (a) Evolution curves for ASI-Evolve, GEPA, and OpenEvolve. (b) Evolution curves for ASI-Evolve using GPT-5-mini and Qwen3-32B. Shaded regions indicate variability across repeated runs.

Figure 4: Sampling algorithm comparison. Evolution curves for ASI-Evolve with MAP-Elites, UCB1, and Random sampling on Qwen3-32B. Shaded regions show run-to-run variability.

Figure 5: Ablation on circle packing. Evolution curves for the full method and ablated variants. Shaded regions indicate variability across repeated runs.

		BindingDB-Dev		BindingDB-Random		Human		BioSNAP
Discovered by	Model	AUROC	F1	AUROC	F1	AUROC	F1	AUROC	F1
Human	TransformerCPI	-	-	93.96	87.02	96.03	91.35	86.35	78.53
	PSICHIC	-	-	91.67	83.30	98.55	94.85	91.64	84.87
	ConPlex	-	-	93.59	83.39	97.04	90.38	88.40	72.86
	ColdStartCPI	-	-	89.27	85.27	98.29	93.57	93.39	86.84
	DrugBAN	94.15	86.89	94.89	87.96	98.61	95.40	89.43	82.69

Table 5: Performance comparison on drug-target interaction prediction across multiple benchmarks. We report AUROC (%) and F1 (%) scores for key evaluation settings. Bold indicates best performance.

		Unseen Drug		Unseen Protein		Unseen Drug and Protein
Discovered by	Model	AUROC	F1	AUROC	F1	AUROC	F1
Human	DrugBAN	79.15	77.39	82.26	75.76	76.47	71.53
AI	Ours	86.09	82.35	85.82	78.44	80.83	74.51

Table 6: Cold-start performance comparison. We report AUROC (%) and F1 (%) scores. Models must predict interactions for unseen drugs, proteins, or both.

Autonomous Research Agents

Autonomous research agents use language models and tools to search literature, design and run experiments, analyze results, and support scientific discovery.

See all 318 papers

Neural Architecture Discovery via Autonomous Evolution
Jul 24, 2025Weixian Xu, Yixiu Liu, Yang Nan +4Autonomous Research Agents AI Agents for Scientific Discovery
Shanghai Jiao Tong University · SII · GAIR +1
ExpEvolve: Structuring and Reusing Experience for Open-Ended Discovery
Oct 21, 2026Yongqiang Chen, Chenxi Liu, Deyu Zou +4AI Agents for Scientific Discovery Large Language Model-Guided Scientific Discovery
MBZUAI · Carnegie Mellon University · TMLR Group, Hong Kong Baptist University +2
FreeEvolve: Learning to Evolve Beyond Fixed Loops
Oct 6, 2026Lecheng Kong, Like Hui, Nikos Kanakaris +3Self-Evolving Agents LLM Agent Self-Improvement
AWS AI Labs · Georgia Institute of Technology

Jul 24, 2025J/K move · Enter open · S save

Neural Architecture Discovery via Autonomous Evolution

Weixian Xu, Yixiu Liu, Yang Nan, Lyumanshan Ye+3

Shanghai Jiao Tong University · SII · GAIR+1

Autonomous Research Agents AI Agents for Scientific Discovery Neural Architecture Search Large Language Model-Guided NAS Linear Attention

Open paper AskPDF arXiv

Recent progress in LLM agents has advanced the prospect of autonomous research. Yet whether AI can complete difficult long-horizon tasks, especially those that advance AI research itself, remains largely unexplored. We present ASI-Arch, a system for AI-driven AI research that autonomously conducts neural architecture research through a closed-loop research-experiment-analyze-update process. Applied to linear attention, ASI-Arch ran 1,773 iterative experiments and discovered 105 state-of-the-art architectures. Its best architecture improves over DeltaNet by nearly three times the gain achieved by Mamba2. Beyond the final performance gains, we analyze the contributions of different parts of the framework in this hard research setting, shedding light on what enables autonomous progress in complex AI research tasks.

Abstract

Autonomous Research Agents

Explore similar work

Neural Architecture Discovery via Autonomous Evolution

ExpEvolve: Structuring and Reusing Experience for Open-Ended Discovery

FreeEvolve: Learning to Evolve Beyond Fixed Loops