Self-Improving AI Agents through Self-Play

• URL: http://arxiv.org/abs/2512.02731v1
• Date: Tue, 02 Dec 2025 13:13:57 GMT
• Title: Self-Improving AI Agents through Self-Play
• Abstract summary: This paper formalizes the AAI capability score as a flow $_r$ parameterized by computational resource $r$.<n>We show that a sufficient condition for $> 0$ is that, up to curvature and step-size effects, the combined noise of generation and verification must be small enough.<n>We apply this formalism to unify the recent literature on Language Self-Play (LSP), Self-Correction, and Synthetic Dataping.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We extend the moduli-theoretic framework of psychometric batteries to the domain of dynamical systems. While previous work established the AAI capability score as a static functional on the space of agent representations, this paper formalizes the agent as a flow $ν_r$ parameterized by computational resource $r$, governed by a recursive Generator-Verifier-Updater (GVU) operator. We prove that this operator generates a vector field on the parameter manifold $Θ$, and we identify the coefficient of self-improvement $κ$ as the Lie derivative of the capability functional along this flow. The central contribution of this work is the derivation of the Variance Inequality, a spectral condition that is sufficient (under mild regularity) for the stability of self-improvement. We show that a sufficient condition for $κ> 0$ is that, up to curvature and step-size effects, the combined noise of generation and verification must be small enough. We then apply this formalism to unify the recent literature on Language Self-Play (LSP), Self-Correction, and Synthetic Data bootstrapping. We demonstrate that architectures such as STaR, SPIN, Reflexion, GANs and AlphaZero are specific topological realizations of the GVU operator that satisfy the Variance Inequality through filtration, adversarial discrimination, or grounding in formal systems.

Related papers

• Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement [23.752667365287568]
  Generalized Agent Iteration (GAI) is a formal framework that describes iterative policy improvement.<n>Two pivotal dials then distinguish the instances: whether the improving mechanism is part of the agent and whether the standard it is measured against is grounded outside it.<n>We see this paper as a first step toward exploring a formal characterization of RSI that rests on the classical account, makes existing systems comparable, and provides a principled basis for analyzing and designing new ones.
  arXiv  Detail & Related papers  (2026-09-11T18:15:09Z)
• Formal Verification of Agentic Systems over Operational Data [59.98246281888422]
  Agentic systems driven by large language models (LLMs) are increasingly deployed in real-world where they act on persistent operational data.<n>Existing approaches do not provide such system-level guarantees.<n>We study here the verification of agentic systems comprising a single LLM and a tool orchestration harness over relational operational data.
  arXiv  Detail & Related papers  (2026-08-04T13:01:30Z)
• Inverting the Bellman Equation: From $Q$-Values to World Models [57.827849584133425]
  We prove that value-based agents trained on a sufficiently rich set of reward functions implicitly encode a unique and accurate world model.<n>We find that policies trained exclusively on a ttReacher agent's implicit world model are quasi-optimal on out-of-distribution, velocity-based goals despite position-only training.
  arXiv  Detail & Related papers  (2026-06-19T07:26:14Z)
• Dynamic analysis enhances issue resolution [53.50448142467294]
  DAIRA (Dynamic Analysis-enhanced Issue Resolution Agent) is an automated repair framework that embeds dynamic analysis into the agent's reasoning cycle.<n> Driven by a Test Tracing-Driven methodology, DAIRA utilizes lightweight monitors to extract critical runtime data.<n>When powered by Gemini 3 Flash Preview, DAIRA establishes a new state-of-the-art (SOTA) performance, achieving a 79.4% resolution rate on the SWE-bench Verified dataset.
  arXiv  Detail & Related papers  (2026-03-23T14:48:54Z)
• Deep Delta Learning [91.75868893250662]
  We introduce Deep Delta Learning (DDL), a novel architecture that generalizes the standard residual connection.<n>We provide a spectral analysis of this operator, demonstrating that the gate $(mathbfX)$ enables dynamic between identity mapping, projection, and geometric reflection.<n>This unification empowers the network to explicitly control the spectrum of its layer-wise transition operator, enabling the modeling of complex, non-monotonic dynamics.
  arXiv  Detail & Related papers  (2026-01-01T18:11:38Z)
• The Geometry of Benchmarks: A New Path Toward AGI [0.0]
  We introduce a geometric framework in which all psychometric batteries for AI agents are treated as points in a structured moduli space.<n>First, we define an Autonomous AI (AAI) Scale, a Kardashev-style hierarchy of autonomy grounded in measurable performance.<n>Second, we construct a moduli space of batteries, identifying equivalence classes of benchmarks that are indistinguishable at the level of agent orderings and capability inferences.<n>Third, we introduce a general Generator-Verifier-Updater (GVU) operator that subsumes reinforcement learning, self-play, debate and verifier-based fine-tuning
  arXiv  Detail & Related papers  (2025-12-03T21:34:09Z)
• Alita-G: Self-Evolving Generative Agent for Agent Generation [54.49365835457433]
  We present ALITA-G, a framework that transforms a general-purpose agent into a domain expert.<n>In this framework, a generalist agent executes a curated suite of target-domain tasks.<n>It attains strong gains while reducing computation costs.
  arXiv  Detail & Related papers  (2025-10-27T17:59:14Z)
• AILoRA: Function-Aware Asymmetric Initialization for Low-Rank Adaptation of Large Language Models [11.663809872664105]
  Low-Rank Adaptation (LoRA) has emerged as one of the most widely adopted approaches.<n>LoRA is typically applied to the $WQ$ and $WV$ projection matrices of self-attention modules.<n>We introduce textAILoRA, a novel parameter-efficient method that incorporates function-aware asymmetric low-rank priors.
  arXiv  Detail & Related papers  (2025-10-09T10:13:16Z)
• Tractable Asymmetric Verification for Large Language Models via Deterministic Replicability [0.6117371161379209]
  The landscape of Large Language Models (LLMs) shifts rapidly towards dynamic, multi-agent systems.<n>This paper proposes a verification framework that achieves tractable asymmetric effort.<n>We show that targeted verification can be over 12 times faster than full regeneration.
  arXiv  Detail & Related papers  (2025-09-14T03:30:06Z)
• HiVA: Self-organized Hierarchical Variable Agent via Goal-driven Semantic-Topological Evolution [13.440964262446558]
  Hierarchical Variable Agent (HiVA) is a novel framework modeling agentic as self-organized graphs with the Semantic-Topological Evolution (STEV) algorithm.<n> Experiments on dialogue, coding, Longcontext Q&A, mathematical, and agentic benchmarks demonstrate improvements of 5-10% in task accuracy and enhanced resource efficiency.
  arXiv  Detail & Related papers  (2025-08-29T18:51:18Z)
• Pruned-ADAPT-VQE: compacting molecular ansätze by removing irrelevant operators [0.0]
  ADAPT-VQE is a problem-tailored variational quantum eigensolver.<n>It selects operators based on their gradient, constructing ans"atze that continuously evolve to match the energy landscape.<n>We propose an automated, cost-free refinement method that removes unnecessary operators from the ansatz.
  arXiv  Detail & Related papers  (2025-04-07T00:54:31Z)
• Semantic Operators: A Declarative Model for Rich, AI-based Data Processing [27.294714926492187]
  We introduce semantic operators, the first formalism for declarative and general-purpose AI-based transformations.<n>We propose several novel optimizations to accelerate semantic filtering, joining, group-by and top-k operations by up to $1,000times$.<n>We show that the semantic operator model is expressive, capturing state-of-the-art AI pipelines in a few operator calls.
  arXiv  Detail & Related papers  (2024-07-16T06:19:14Z)
• Delta-AI: Local objectives for amortized inference in sparse graphical models [64.5938437823851]
  We present a new algorithm for amortized inference in sparse probabilistic graphical models (PGMs) Our approach is based on the observation that when the sampling of variables in a PGM is seen as a sequence of actions taken by an agent, sparsity of the PGM enables local credit assignment in the agent's policy learning objective. We illustrate $Delta$-AI's effectiveness for sampling from synthetic PGMs and training latent variable models with sparse factor structure.
  arXiv  Detail & Related papers  (2023-10-03T20:37:03Z)
• Disentanglement via Latent Quantization [60.37109712033694]
  In this work, we construct an inductive bias towards encoding to and decoding from an organized latent space. We demonstrate the broad applicability of this approach by adding it to both basic data-re (vanilla autoencoder) and latent-reconstructing (InfoGAN) generative models.
  arXiv  Detail & Related papers  (2023-05-28T06:30:29Z)
• Learning Dynamic Mechanisms in Unknown Environments: A Reinforcement Learning Approach [123.55983746427572]
  We propose novel learning algorithms to recover the dynamic Vickrey-Clarke-Grove (VCG) mechanism over multiple rounds of interaction.<n>A key contribution of our approach is incorporating reward-free online Reinforcement Learning (RL) to aid exploration over a rich policy space.
  arXiv  Detail & Related papers  (2022-02-25T16:17:23Z)
• Utilizing Redundancy in Cost Functions for Resilience in Distributed Optimization and Learning [1.8414221462731502]
  This paper considers the problem of resilient distributed optimization and machine learning in a server-based architecture. The system comprises a server and multiple agents, where each agent has a local cost function. We consider the case when some of the agents may be asynchronous and/or Byzantine faulty.
  arXiv  Detail & Related papers  (2021-10-21T02:41:19Z)
• Implicit Distributional Reinforcement Learning [61.166030238490634]
  implicit distributional actor-critic (IDAC) built on two deep generator networks (DGNs) Semi-implicit actor (SIA) powered by a flexible policy distribution. We observe IDAC outperforms state-of-the-art algorithms on representative OpenAI Gym environments.
  arXiv  Detail & Related papers  (2020-07-13T02:52:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.