Most contemporary artificial intelligence (AI) systems are released with frozen parameters and improve through external updates, whereas an ideal system would keep improving from its own experience, a capacity we call recursive self-improvement (RSI). Yet studying RSI requires an environment that is continuously changing, objectively verifiable, and naturally grounded, posing a fundamental challenge for agent evolution. Here we introduce EvoForesight, a sustainable evolution agent system grounded in future event prediction, in which agents continuously interact with a dynamically evolving environment and improve by internalizing and externalizing the experience gained through these interactions. A living environment synthesizes verifiable tasks, and a recursive-evolution engine turns the reasoning trajectory behind each prediction back into capability: decoupling search from prediction adapts parameter-level evolution to the contamination-prone live environment, while distilling reusable skills lets the agent evolve from the limited live trajectories such an environment affords. On the external FutureX leaderboard, which evaluates predictions against real-world outcomes and is updated weekly, our self-improved forecaster ranked first across all systems over three consecutive weeks, achieving up to a 195% improvement in rating. On our synthesized evaluation, it ranked among the top three across four base LLMs, outperforming the strongest baseline by up to 10.7 Bradley--Terry points. More broadly, these results suggest that a system built for self-improvement is best understood as an agent that co-evolves with a genuinely living, open-ended environment, pointing to a shift in AI from training static models on fixed data toward cultivating agents that grow stronger from their own experience, with such an open world as the indispensable basis of that growth.
Figures & tables
Figure 1: Overview of EvoForesight for recursive self-improvement in real-world event forecasting. a, The closed loop connects question generation, forecasting, and evolution, with resolved outcomes supplying verifiable rewards and updates feeding into subsequent rounds. b, Prometheus generates questions through templates and agent-driven discovery, followed by temporal validation, deduplication, conflict detection, and diversity-aware selection. c, Branchus separates evidence gathering, by a search team of a scheduler, searcher, and critic, from probabilistic prediction, by a predict team . The search team is time-sensitive, as its result depends on when it runs, while the predict team is time-invariant, reasoning only over the search team’s cached evidence and reasoning trajectories. Both teams retrieve relevant forecasting skills. d, Athena updates the predict team’s parameters using cached evidence and verifiable rewards, distills search trajectories into reusable forecasting skills, and converts observed weaknesses into task-synthesis skills for Prometheus . Resolved outcomes inform parameter and environment updates but are withheld from the behavior-level judge.
Figure 2: Overall FutureX ranking across four weekly evaluations. Each panel reports the sliding-window Bradley–Terry rating of every harness for one week, with Branchus (ours) shown against the baseline harnesses. The complete official leaderboards are provided in Appendix A (Figure 11 ).
Harness
Choice Questions
Numeric Questions
Overall Performance
Accuracy ↑
Brier ↓
AUROC ↑
Relative WIS ↓
BT ↑
Rank ↓
Qwen3.8-Max
ReAct
70.7
0.346
0.784
0.913
47.9
16
MAS
70.3
0.343
0.808
0.871
51.5
11
OpenClaw
71.9
0.356
0.749
0.933
47.1
18
DeerFlow
71.9
0.354
0.748
0.927
45.5
20
Table 1: Forecasting performance of Branchus against baseline harnesses across four LLMs on the Prometheus -synthesized evaluation. We report accuracy, Brier, and AUROC for choice questions, relative WIS for numeric questions, and the Bradley–Terry score (BT) with its rank overall. The rank is a single ordering over all 23 harness–LLM systems in the table, not a within-block rank . Best values per column within each base-LLM block are in bold .
Figure 3: Parameter evolution improves forecasting quality continuously. Across two evolution rounds, the training set grows from 600 to 1,000 examples, accompanied by consistent improvements in forecasting performance. Left : the performance of relative WIS. Right : calibration, measured by the Brier score. For both metrics, lower is better.
Figure 4: RFT delivers consistent, calibration-led gains over the base model. Each panel traces the Base → RFT transition for two LLMs on one axis of forecasting quality, with every axis oriented so that upward movement is an improvement. Fine-tuning on successful trajectories lifts every axis, and the improvement is led by calibration rather than raw accuracy. Left : choice accuracy. Right : numeric interval forecasts, measured by relative WIS.
Figure 5: RLVR sharpens both forecasting-quality metrics across LLMs. For each LLM, we compare the base model against its RLVR-tuned counterpart on choice accuracy (left blue axis) and the numeric score (right red axis, 1−bWIS ). Each metric shows two side-by-side bars, the base model and its RLVR-tuned counterpart, as absolute values. RLVR improves both metrics on both models.
Figure 6: Evolving skills improve forecasting quality continuously. Distilled skills, inserted at inference without touching the weights, sharpen Branchus ’s forecasts on both axes of forecasting quality. Because each round adds its live trajectories to the pool (nested pools of 88 , 186 , and 262 trajectories), the skills Athena extracts over successive evolution rounds lift forecasting quality round by round rather than in a one-off gain. Left : numeric accuracy, measured by MAPE. Right : calibration, measured by the Brier score. For both metrics, lower is better.
Figure 7: Distilled skills scale across evolution rounds. As the collected trajectories grow round by round, Athena keeps distilling new skills from them, so the library expands rather than saturating, and later skills extend into domains seen only rarely before. Left : the number of distilled skills at each evolution round. Right : the breadth of topical coverage, measured by the number of the domains and concrete tasks that the skills cover.
Figure 8: Skills versus parameter updates: where they help and what they cost. Externalizing experience into skills and internalizing it into parameters act at different stages and carry different costs, so the two axes are complementary rather than interchangeable. Left : forecasting accuracy and the agent’s tool-call budget as the skill library grows, separated into structure synthesis problems that reuse recurring patterns and open synthesis problems from unfamiliar domains. Right : the compute cost of skill extraction against parameter updates as the base model scales up.
Stage
Step
Calls
Input
Output
Total
Evaluation
Step scoring (searcher, critic)
6.49
100,781
8,476
109,257
Scheduler scoring
1.45
34,487
2,259
36,745
Trajectory diagnosis
1.05
52,396
1,098
53,494
Extraction
Experience extraction
1.02
29,363
1,372
30,735
Experience filtering
0.91
29,520
508
30,029
Generalization
Four-level generalization
2.02
18,409
2,354
20,763
Table 2: Token cost of distilling skills from one trajectory, by stage. Values are averages per trajectory; output tokens include reasoning tokens.
Figure 9: Environment evolution targets genuine weaknesses and enriches the resulting experience. Coupling task synthesis to the agent’s own failure modes turns its weaknesses into harder challenges, and the richer trajectories these produce feed back into the system’s evolution. Left : the difficulty of the synthesized questions under the original versus the evolved skill, showing that evolution raises difficulty while keeping each question well-posed and resolvable. Right : the distilled experience obtained from the same trajectory pool, comparing a weakness-targeting environment against a fixed one in the number of skills and the breadth of their coverage.
Figure 10: Fine-tuning on the harder, evolved questions pays off. Across every metric, RFT on the adversarially synthesized hard problems improves over plain RFT. Each row is drawn on its own scale.
Appendix figures & tables4 assets
Supplementary material from the paper’s appendix.
Appendix
Figure 11: Full FutureX overall leaderboards for the four weekly evaluations. Official leaderboard screenshots showing the complete field of harnesses ranked by sliding-window Bradley–Terry rating. These correspond, in order, to the four panels of Figure 2 .
Prize
Published forecast
Physics
x.com/BranchusAI/status/2107395342539030965
Chemistry
x.com/BranchusAI/status/2107756760409034917
Literature
x.com/BranchusAI/status/2108101909391524193
Peace
x.com/BranchusAI/status/2108460181793419444
Economics
x.com/BranchusAI/status/2109501869265658231
Appendix
Table 3: Pre-announcement forecasts for the 2026 Nobel Prizes, published on X before each laureate was announced.
Figure 12: Pre-announcement forecasts for the 2026 Nobel Prizes in Physics and Peace. Each panel ranks forecasters by total Brier score (lower is better) against the no-forecast baseline of 1.0 . Both laureates were outsiders that no forecaster made a favourite, so no system beat the baseline. Branchus was the only forecaster to list either laureate, surfacing Halzen in Physics and Pillay in Peace in one of its three independent runs.
Figure 13: Per-candidate pre-announcement forecasts for the 2026 Nobel Prizes in Literature and Economics. Each cell is the probability a forecaster assigned to a candidate being among the laureates, and bold marks each column’s top candidate.
Self-evolving agents improve through repeated cycles of modification and evaluation. However, existing methods receive feedback on modifications only after costly evaluations, and aggregate scores may obscure distinctions between failures that require different repairs. We introduce predictive self-evolution through an editable Improvement Dynamics Model (IDM). Before evaluation, the IDM predicts task-level outcomes of a proposed modification and guides an optional revision. After evaluation, past modifications and their observed outcomes are used to improve the IDM, allowing predictive knowledge to accumulate across generations. The system can thus improve how it solves tasks, proposes modifications, and predicts their effects. Building on HyperAgents, we evaluate this approach across multilingual coding, paper review, robotics reward design, and Olympiad-level mathematical grading. Our approach outperforms HyperAgents across all four domains, with up to 46% higher average performance over generations. Gains also persist under matched token budgets that include both agent evolution and task evaluation, and extend to held-out tasks. Forecast accuracy improves as the IDM evolves, and the learned predictive knowledge transfers to improve evolution in a new domain.
Zixuan Xie, Chang Xu, Yuhua Jiang +1
University of Virginia · Microsoft Research · Tsinghua University
Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input. This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains. We offer a system-level framework that represents a modern agent as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic. Within this framework, self-improvement is formalized as a self-induced update operator that obtains and commits updates to model parameters or scaffold components. We organize prior work by update target and by the signals that drive change, then review applications and discuss evaluation, before closing with open problems and future directions. For convenience, we track technical updates on https://github.com/selfimproving-agent/awesome-Self-Improving-Agents.
Zhe Ren, Yimeng Chen, Dandan Guo +9
School of Artificial Intelligence, Jilin University · King Abdullah University of Science and Technology (KAUST) · Independent Researcher +2
Recursive self-improvement (RSI) offers a route to AI systems that participate in their own continued development. Recent methods allow language agents to retain changes to their parameters, memory, skills, and code for subsequent execution and improvement. We examine how these methods can inform new research ideas through two complementary perspectives. First, we introduce a unified taxonomy with nine design dimensions organized into four groups: update targets, update organization, update schedule and scope, feedback and selection. This taxonomy places existing methods in a shared design space, where examining individual dimensions and their interactions can inspire new research ideas. Second, we examine how existing RSI methods draw on traditional deep learning, for example through optimization, regularization, data augmentation, continual learning, and adversarial learning. We develop further ideas by examining more specific learning mechanisms and drawing on evolutionary computation. We also review how existing benchmarks assess self-improvement under online, offline, and offline-to-online settings. Together, these perspectives help researchers develop new RSI research ideas from existing designs and learning mechanisms. RSI papers and resources are available at https://prism-shadow.github.io/awesome-rsi/.
Leijun Zhou, Yanke Yu, Zhihao Liu +10
PrismShadow · Peking University · Zhejiang University +1