Self-Improving Agents

Ask about this topic

Overview

Self-improving agents study how systems can learn from their own actions and outcomes to perform better on later tasks, from coding and web use to robotics and long-horizon games. Common approaches turn successes and failures into reusable memories or executable skills, revise prompts and agent harnesses, or update model parameters through fine-tuning and reinforcement learning. Work increasingly uses repeated trials, held-out tasks, execution-based checks, and external evaluators to test whether improvements transfer rather than merely exploit a score. The central tension is that feedback and self-authored tests can be unreliable, so genuine progress must be balanced against regression, reward hacking, safety risks, and the cost of verification.

Latest papers

CardsList

171 papers57 papers in the last four weeks, up 171% on the four weeks before. 0.3% of all new papers.

Apr 8, 2026

PRIME: Training Free Proactive Reasoning via Iterative Memory Evolution for User-Centric Agent

The development of autonomous tool-use agents for complex, long-horizon tasks in collaboration with human users has become the frontier of agentic research. During multi-turn Human-AI interactions, the dynamic and uncertain nature of user demands poses a significant challenge; agents must not only invoke tools but also iteratively refine their understanding of user intent through effective communication. While recent advances in reinforcement learning offer a path to more capable tool-use agents, existing approaches require expensive training costs and struggle with turn-level credit assignment across extended interaction horizons. To this end, we introduce PRIME (Proactive Reasoning via Iterative Memory Evolution), a gradient-free learning framework that enables continuous agent evolvement through explicit experience accumulation rather than expensive parameter optimization. PRIME distills multi-turn interaction trajectories into structured, human-readable experiences organized across three semantic zones: successful strategies, failure patterns, and user preferences. These experiences evolve through meta-level operations and guide future agent behavior via retrieval-augmented generation. Our experiments across several diverse user-centric environments demonstrate that PRIME achieves competitive performance with gradient-based methods while offering cost-efficiency and interpretability. Together, PRIME presents a practical paradigm for building proactive, collaborative agents that learn from Human-AI interaction without the computational burden of gradient-based training.

Prince Zizhuang Wang, Shuli Jiang

Carnegie Mellon University

Self-Improving AgentsMemory-Augmented Language Model AgentsContinual Learning for LLM AgentsHuman-AI InteractionTest-Time Training
AskPDF
Apr 2, 2026

The Self Driving Portfolio: Agentic Architecture for Institutional Asset Management

Agentic AI shifts the investor's role from analytical execution to oversight. We present an agentic strategic asset allocation pipeline in which 44 specialized agents produce capital market assumptions, construct portfolios using 21 competing methods, and critique and vote on each other's outputs. A researcher agent proposes new portfolio construction methods not yet represented, and a meta agent compares past forecasts against realized returns and rewrites agent code and prompts to improve future performance. The entire pipeline is governed by the Investment Policy Statement - the same document that guides human portfolio managers can now constrain and direct autonomous agents.

Andrew Ang, Nazym Azimbayev, Andrey Kim

Tau Balance and an adjunct professor at Columbia University · Altbridge

Financial ServicesMulti-Agent OrchestrationPortfolio OptimizationAI Agent GovernanceSelf-Improving Agents
AskPDF
Apr 1, 2026

PsychAgent: An Experience-Driven Lifelong Learning Agent for Self-Evolving Psychological Counselor

Existing methods for AI psychological counselors predominantly rely on supervised fine-tuning using static dialogue datasets. However, this contrasts with human experts, who continuously refine their proficiency through clinical practice and accumulated experience. To bridge this gap, we propose an Experience-Driven Lifelong Learning Agent (PsychAgent) for psychological counseling. First, we establish a Memory-Augmented Planning Engine tailored for longitudinal multi-session interactions, which ensures therapeutic continuity through persistent memory and strategic planning. Second, to support self-evolution, we design a Skill Evolution Engine that extracts new practice-grounded skills from historical counseling trajectories. Finally, we introduce a Reinforced Internalization Engine that integrates the evolved skills into the model via rejection fine-tuning, aiming to improve performance across diverse scenarios. Comparative analysis shows that our approach achieves higher scores than strong general LLMs (e.g., GPT-5.4, Gemini-3) and domain-specific baselines across all reported evaluation dimensions. These results suggest that lifelong learning can improve the consistency and overall quality of multi-session counseling responses.

Yutao Yang, Junsong Li, Qianjun Pan +7

School of Computer Science and Technology, East China Normal University · Shanghai AI Laboratory · School of Psychology and Cognitive Science, East China Normal University

Lifelong Learning AgentsAgent Skill LearningSelf-Improving AgentsMental Health CounselingMemory-Augmented Language Model Agents
AskPDF
Apr 1, 2026

Meta-TTL: Meta-Learning Self-Improvement Policies for Language Agents

Test-Time Learning (TTL) enables language agents to iteratively refine their performance through repeated interactions with the environment at inference time. At the core of TTL is a self-improvement policy that updates the actor policy based on experience from previous episodes, thereby improving future behavior. Existing methods rely on hand-crafted self-improvement rather than optimizing them for downstream improvement. We argue that optimal self-improvement policies should be learned from task environments, not hand-engineered based on human intuition. To achieve this, we introduce Meta-TTL, a framework that formulates the discovery of effective self-improvement policies as a bi-level optimization problem. Within this framework, the inner loop executes the standard TTL process, measuring how effectively a candidate self-improvement policy helps an agent correct errors across sequential episodes. Guided by the agent's performance, the outer loop performs reflective meta-training across diverse training tasks, using a balanced improvement score (BIS) to balance task contributions during candidate selection. We evaluate Meta-TTL on Jericho, WebArena-Lite, and -bench across both in-distribution (ID) and out-of-distribution (OOD) settings. Meta-TTL consistently outperforms existing baselines, improving TTL over the strongest baseline by up to 23% on ID tasks and 27% on OOD tasks. These results suggest that the optimized self-improvement policy encodes transferable meta-strategies that generalize beyond the training task distribution.

Zhanzhi Lou, Hui Chen, Yibo Li +2

National University of Singapore

Meta-LearningTest-Time TrainingLLM Agent TrainingSelf-Improving Agents
AskPDF
Mar 31, 2026

Empirical Validation of the Classification-Verification Dichotomy for AI Safety Gates

Can classifier-based safety gates maintain reliable oversight as AI systems improve over hundreds of iterations? We provide comprehensive empirical evidence that they cannot. On a self-improving neural controller (d=240), eighteen classifier configurations -- spanning MLPs, SVMs, random forests, k-NN, Bayesian classifiers, and deep networks -- all fail the dual conditions for safe self-improvement. Three safe RL baselines (CPO, Lyapunov, safety shielding) also fail. Results extend to MuJoCo benchmarks (Reacher-v4 d=496, Swimmer-v4 d=1408, HalfCheetah-v4 d=1824). At controlled distribution separations up to delta_s=2.0, all classifiers still fail -- including the NP-optimal test and MLPs with 100% training accuracy -- demonstrating structural impossibility. We then show the impossibility is specific to classification, not to safe self-improvement itself. A Lipschitz ball verifier achieves zero false accepts across dimensions d in {84, 240, 768, 2688, 5760, 9984, 17408} using provable analytical bounds (unconditional delta=0). Ball chaining enables unbounded parameter-space traversal: on MuJoCo Reacher-v4, 10 chains yield +4.31 reward improvement with delta=0; on Qwen2.5-7B-Instruct during LoRA fine-tuning, 42 chain transitions traverse 234x the single-ball radius with zero safety violations across 200 steps. A 50-prompt oracle confirms oracle-agnosticity. Compositional per-group verification enables radii up to 37x larger than full-network balls. At d<=17408, delta=0 is unconditional; at LLM scale, conditional on estimated Lipschitz constants.

Arsenios Scrivens

AI Agent SafetyAI Safety EvaluationRuntime Enforcement for AI AgentsSelf-Improving AgentsNeural Network Verification
AskPDF
Mar 25, 2026

Experiential Reflective Learning for Self-Improving LLM Agents

Recent advances in large language models (LLMs) have enabled the development of autonomous agents capable of complex reasoning and multi-step problem solving. However, these agents struggle to adapt to specialized environments and do not leverage past interactions, approaching each new task from scratch regardless of their accumulated experience. We introduce Experiential Reflective Learning (ERL), a simple self-improvement framework that enables rapid environment adaptation through experiential learning. ERL reflects on task trajectories and outcomes to generate heuristics, capturing actionable lessons that transfer across tasks. At test time, relevant heuristics are retrieved based on the current task and injected into the agent's context to guide execution. On the Gaia2 benchmark, ERL improves success rate by 7.8% over a ReAct baseline, with large gains in task completion reliability, and outperforms prior experiential learning methods. Through systematic ablations, we find that selective retrieval is essential and that heuristics provide more transferable abstractions than few-shot trajectory prompting. These results demonstrate that reflecting on single-attempt experiences to extract transferable heuristics enables effective agent self-improvement.

Marc-Antoine Allard, Arnaud Teinturier, Victor Xing +1

Illuin Technology

Self-Improving AgentsLLM Agent TrainingMemory-Augmented Language Model Agents
AskPDF
Mar 25, 2026

Understanding the Challenges in Iterative Generative Optimization with LLMs

Generative optimization uses large language models (LLMs) to iteratively improve artifacts (such as code, workflows or prompts) using execution feedback. It is a promising approach to building self-improving agents, yet in practice remains brittle: despite active research, only 9% of surveyed agents used any automated optimization. We argue that this brittleness arises because, to set up a learning loop, an engineer must make ``hidden'' design choices: What can the optimizer edit and what is the "right" learning evidence to provide at each update? We investigate three factors that affect most applications: the starting artifact, the credit horizon for execution traces, and batching trials and errors into learning evidence. Through case studies in MLAgentBench, Atari, and BigBench Extra Hard, we find that these design decisions can determine whether generative optimization succeeds, yet they are rarely made explicit in prior work. Different starting artifacts determine which solutions are reachable in MLAgentBench, truncated traces can still improve Atari agents, and larger minibatches do not monotonically improve generalization on BBEH. We conclude that the lack of a simple, universal way to set up learning loops across domains is a major hurdle for productionization and adoption. We provide practical guidance for making these choices.

Allen Nie, Xavier Daull, Zhiyi Kuang +10

Google DeepMind · CNRS · Stanford University +5

Large Language Model-Guided OptimizationSelf-Improving Agents
AskPDF
Mar 24, 2026

Bilevel Autoresearch: Meta-Autoresearching Itself

If autoresearch is itself a form of research, then autoresearch can be applied to research itself. We present Bilevel Autoresearch, a bilevel framework in which an outer autoresearch loop improves an inner autoresearch loop by reading its code and traces, identifying bottlenecks, and generating injectable Python search mechanisms at runtime. The inner loop optimizes task performance; the outer loop optimizes how the inner loop searches. Both loops use the same LLM, so improvements come from the bilevel architecture rather than a stronger meta-level model, although the outer loop consumes additional inference and wall-clock budget. On Karpathy's GPT pretraining benchmark, the meta-autoresearch outer loop achieves a 5x improvement over the standard inner loop alone (-0.045 vs. -0.009 val_bpb), while parameter-level adjustment without mechanism change yields no reliable gain. The outer loop instantiates mechanisms from adjacent search domains, including combinatorial optimization, multi-armed bandits, and design of experiments, without human specification of the final mechanism design. Trace analysis suggests that these mechanisms break deterministic search patterns and force exploration of directions the LLM's priors avoid. The experiments demonstrate, on this benchmark, a first bilevel step: an outer loop improves the search behavior of an inner loop. Code is the mechanism carrier in this implementation, but skills, prompts, workflows, evaluators, domain principles, world-model assumptions, and memory schemas can also encode mechanisms that shape future agent behavior. This suggests a path toward recursive bootstrapping, where mechanisms discovered for the inner loop can be fed back to improve the meta-level loop itself.

Yaonan Qu, Meng Lu

Independent Researcher

Self-Improving AgentsAutonomous Research AgentsLarge Language Model-Guided OptimizationAutomated Algorithm DiscoveryBilevel Optimization
AskPDF
Mar 24, 2026

Polaris: A Gödel Agent Framework for Small Language Models through Experience-Abstracted Policy Repair

Gödel agent realize recursive self-improvement: an agent inspects its own policy and traces and then modifies that policy in a tested loop. We introduce Polaris, Gödel agent for compact models that performs policy repair via experience abstraction, turning failures into policy updates through a structured cycle of analysis, strategy formation, abstraction, and minimal code patch repair with conservative checks. Unlike response level self correction or parameter tuning, Polaris makes policy level changes with small, auditable patches that persist in the policy and are reused on unseen instances within each benchmark. As part of the loop, the agent engages in meta reasoning: it explains its errors, proposes concrete revisions to its own policy, and then updates the policy. To enable cumulative policy refinement, we introduce experience abstraction, which distills failures into compact, reusable strategies that transfer to unseen instances. On MGSM, DROP, GPQA, and LitBench (covering arithmetic reasoning, compositional inference, graduate-level problem solving, and creative writing evaluation), a 7-billion-parameter model equipped with Polaris achieves consistent gains over the base policy and competitive baselines.

Aditya Kakade, Vivek Srivastava, Shirish Karande

TCS Research, India

Small Language ModelsSelf-Improving AgentsRecursive Self-Improvement
AskPDF
Mar 19, 2026

Hyperagents

Self-improving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problem-solving processes. Existing approaches to self-improvement rely on fixed, handcrafted meta-level mechanisms, fundamentally limiting how fast such systems can improve. The Darwin Gödel Machine (DGM) demonstrates open-ended self-improvement in coding by repeatedly generating and evaluating self-modified variants. Because both evaluation and self-modification are coding tasks, gains in coding ability can translate into gains in self-improvement ability. However, this alignment does not generally hold beyond coding domains. We introduce hyperagents, self-referential agents that integrate a task agent (which solves the target task) and a meta agent (which modifies itself and the task agent) into a single editable program. Crucially, the meta-level modification procedure is itself editable, enabling metacognitive self-modification, improving not only the task-solving behavior, but also the mechanism that generates future improvements. We instantiate this framework by extending DGM to create DGM-Hyperagents (DGM-H), eliminating the assumption of domain-specific alignment between task performance and self-modification skill to potentially support self-accelerating progress on any computable task. Across diverse domains, the DGM-H improves performance over time and outperforms baselines without self-improvement or open-ended exploration, as well as prior self-improving systems. Furthermore, the DGM-H improves the process by which it generates new agents (e.g., persistent memory, performance tracking), and these meta-level improvements transfer across domains and accumulate across runs. DGM-Hyperagents offer a glimpse of open-ended AI systems that do not merely search for better solutions, but continually improve their search for how to improve.

Jenny Zhang, Bingchen Zhao, Wannan Yang +5

University of British Columbia · Vector Institute · University of Edinburgh +4

Self-Improving AgentsRecursive Self-ImprovementSelf-Evolving Agents
AskPDF
Mar 18, 2026

Adaptive Workflow Intelligence: A Cognitive Architecture for Context-Driven Enterprise Automation

Enterprise systems increasingly rely on automated workflows, yet many AI-driven solutions remain brittle under non-stationary conditions, evolving policies, and delayed operational feedback. While reinforcement learning and large language model (LLM) agents offer partial adaptability, they do not by themselves provide persistent reflection mechanisms or straightforward integration with policy-constrained enterprise operations. This paper introduces Adaptive Workflow Intelligence (AWI), a cognitive architecture for context-driven enterprise agents organized around a four-layer Perception-Cognition-Action-Reflection (PCAR) loop. AWI treats reflection as a mechanism for continuous policy refinement and combines hybrid reasoning with reflective memory and feedback-driven adaptation to support decision making under environmental drift and operational constraints. We evaluate AWI in a simulated enterprise decision workflow characterized by delayed outcomes and a controlled regime shift. In a drift-and-delay stress test, guardrail-constrained adaptive approaches recover more rapidly than static automation while maintaining policy compliance. Within this setting, AWI's reflective components modestly reduce behavioral oscillation and feedback variance, illustrating the stability-agility trade-off introduced by reflective policy adaptation.

Sreedevi Pandiyath Viswambaran

Georgia Institute of Technology

Cognitive Architectures for AI AgentsWorkflow AutomationBusiness Process AutomationSelf-Improving Agents
AskPDF
Mar 17, 2026

Internalizing Agency from Reflective Experience

Large language models are increasingly deployed as autonomous agents that must plan, act, and recover from mistakes through long-horizon interaction with environments that provide rich feedback. However, prevailing outcome-driven post-training methods (e.g., RL with verifiable rewards) primarily optimize final success signals, leaving rich environment feedback underutilized. Consequently, they often lead to distribution sharpening: the policy becomes better at reproducing a narrow set of already-successful behaviors, while failing to improve the feedback-grounded agency needed to expand problem-solving capacity (e.g., Pass@k) in long-horizon settings. To address this, we propose LEAFE (Learning Feedback-Grounded Agency from Reflective Experience), a framework that internalizes recovery agency from reflective experience. Specifically, during exploration, the agent summarizes environment feedback into actionable experience, backtracks to earlier decision points, and explores alternative branches with revised actions. We then distill these experience-guided corrections into the model through supervised fine-tuning, enabling the policy to recover more effectively in future interactions. Across a diverse set of interactive coding and agentic tasks under fixed interaction budgets, LEAFE consistently improves Pass@1 over the base model and achieves higher Pass@k than outcome-driven baselines (GRPO) and experience-based methods such as Early Experience, with gains of up to 14% on Pass@128.

Rui Ge, Yichao Fu, Yuyang Qian +4

University of California San Diego · Shanghai Jiao Tong Univsersity · Nanjing University

LLM Agent TrainingSelf-Improving Agents
AskPDF
Mar 12, 2026

HCP-DCNet: A Hierarchical Causal Primitive Dynamic Composition Network for Self-Improving Causal Understanding

The ability to understand and reason about cause and effect -- encompassing interventions, counterfactuals, and underlying mechanisms -- is a cornerstone of robust artificial intelligence. While deep learning excels at pattern recognition, it fundamentally lacks a model of causality, making systems brittle under distribution shifts and unable to answer ``what-if'' questions. This paper introduces the Hierarchical Causal Primitive Dynamic Composition Network (HCP-DCNet), a unified framework that bridges continuous physical dynamics with discrete symbolic causal inference. Departing from monolithic representations, HCP-DCNet decomposes causal scenes into reusable, typed causal primitives organized into four abstraction layers: physical, functional, event, and rule. A dual-channel routing network dynamically composes these primitives into task-specific, fully differentiable Causal Execution Graphs (CEGs). Crucially, the system employs a causal-intervention-driven meta-evolution strategy, enabling autonomous self-improvement through a constrained Markov decision process. We establish rigorous theoretical guarantees, including type-safe composition, routing convergence, and universal approximation of causal dynamics. Extensive experiments across simulated physical and social environments demonstrate that HCP-DCNet significantly outperforms state-of-the-art baselines in causal discovery, counterfactual reasoning, and compositional generalization. This work provides a principled, scalable, and interpretable architecture for building AI systems with human-like causal abstraction and continual self-refinement capabilities.

Ming Lei, Shufan Wu, Christophe Baehr

Shanghai JiaoTong University, Shanghai · M´et´eo France/CNRS, Universit´e de Toulouse, CNRM UMR 3589, Toulouse, France

Causal ReasoningCausal AbstractionCausal Representation LearningCausal DiscoveryCounterfactual ReasoningSelf-Improving Agents
AskPDF
Mar 11, 2026

Reflection-Driven Self-Improving Language Model Agents for Adaptive RAN Slicing

Adaptive control in AI-native networks remains challenging because tasks such as dynamic Radio Access Network (RAN) slicing require persistent trade-off decisions among spectrum efficiency, service quality, and reconfiguration stability. Conventional Reinforcement Learning (RL) compresses these competing objectives into handcrafted scalar rewards, requiring exhaustive tuning for a balanced trade-off. Language model (LM) agents provide a new path toward adaptive control because they can semantically represent complex preference relations, but prompt-based adaptation struggles to retain long-horizon interaction history due to finite context limits. We propose a Reflection-Driven Self-Improvement (ReDSI) framework that replaces scalar rewards with semantic feedback for continuous policy adaptation. This framework is built upon a Small Language Model (SLM) agent for fast decision-making, coupled with a bi-perspective reflection to transform interaction history into preference data, subsequently distilling long-horizon decision experience into the SLM parameters through preference-based fine-tuning. Evaluated on a dynamic RAN slicing task, our method achieves an optimal trade-off across multiple competing objectives, outperforming standard RL baselines and existing LM-based agents by demonstrating substantially greater sample efficiency. These results highlight a viable paradigm of self-improving LM agents for reward-free multi-objective continuous control in AI-native networks.

Yuanhao Li, Haozhe Wang, Geyong Min +2

Department of Computer Science, Faculty of Environment, Science and Economy, University of Exeter, Exeter, EX4 4QF, UK. · British Telecom, UK.

Self-Improving AgentsLanguage Model-Based ControlContinual Learning for LLM AgentsFine-Tuning
AskPDF
Mar 10, 2026

EPOCH: An Agentic Protocol for Multi-Round System Optimization

Autonomous agents are increasingly used to improve prompts, code, and machine learning systems through iterative execution and feedback. Yet existing approaches are usually designed as task-specific optimization loops rather than as a unified protocol for establishing baselines and managing tracked multi-round self-improvement. We introduce EPOCH, an engineering protocol for multi-round system optimization in heterogeneous environments. EPOCH organizes optimization into two phases: baseline construction and iterative self-improvement. It further structures each round through role-constrained stages that separate planning, implementation, and evaluation, and standardizes execution through canonical command interfaces and round-level tracking. This design enables coordinated optimization across prompts, model configurations, code, and rule-based components while preserving stability, reproducibility, traceability, and integrity of evaluation. Empirical studies in various tasks illustrate the practicality of EPOCH for production-oriented autonomous improvement workflows.

Zhanlin Liu, Yitao Li, Munirathnam Srikanth

LLM Agent Workflow OptimizationSelf-Improving AgentsAgent Harness Optimization
AskPDF
Feb 27, 2026

Experience-Guided Self-Adaptive Cascaded Agents for Breast Cancer Screening and Diagnosis with Reduced Biopsy Referrals

We propose an experience-guided cascaded multi-agent framework for Breast Ultrasound Screening and Diagnosis, called BUSD-Agent, that aims to reduce diagnostic escalation and unnecessary biopsy referrals. Our framework models screening and diagnosis as a two-stage, selective decision-making process. A lightweight screening clinic' agent, restricted to classification models as tools, selectively filters out benign and normal cases from further diagnostic escalation when malignancy risk and uncertainty are estimated as low. Cases that have higher risks are escalated to the diagnostic clinic' agent, which integrates richer perception and radiological description tools to make a secondary decision on biopsy referral. To improve agent performance, past records of pathology-confirmed outcomes along with image embeddings, model predictions, and historical agent actions are stored in a memory bank as structured decision trajectories. For each new case, BUSD-Agent retrieves similar past cases based on image, model response and confidence similarity to condition the agent's current decision policy. This enables retrieval-conditioned in-context adaptation that dynamically adjusts model trust and escalation thresholds from prior experiences without parameter updates. Evaluation across 10 breast ultrasound datasets shows that the proposed experience-guided workflow reduces diagnostic escalation in BUSD-Agent from 84.95% to 58.72% and overall biopsy referrals from 59.50% to 37.08%, compared to the same architecture without trajectory conditioning, while improving average screening specificity by 68.48% and diagnostic specificity by 6.33%.

Pramit Saha, Mohammad Alsharid, Joshua Strong +1

Department of Engineering Science, University of Oxford · Department of Computer Science, Khalifa University

Clinical Decision SupportSelf-Improving AgentsMulti-Agent SystemsSelective PredictionUltrasound Imaging
AskPDF
Feb 26, 2026

ParamMem: Augmenting Language Agents with Parametric Reflective Memory

Self-reflection enables language agents to iteratively refine solutions, yet often produces repetitive outputs that limit reasoning performance. Recent studies have attempted to address this limitation through various approaches, among which increasing reflective diversity has shown promise. Our empirical analysis reveals a strong positive correlation between reflective diversity and task success, further motivating the need for diverse reflection signals. We introduce ParamMem, a parametric memory module that encodes cross-sample reflection patterns into model parameters, enabling diverse reflection generation through temperature-controlled sampling. Building on this module, we propose ParamAgent, a reflection-based agent framework that integrates parametric memory with episodic and cross-sample memory. Extensive experiments on code generation, mathematical reasoning, and multi-hop question answering demonstrate consistent improvements over state-of-the-art baselines. Further analysis reveals that ParamMem is sample-efficient, enables weak-to-strong transfer across model scales, and supports self-improvement without reliance on stronger external model, highlighting the potential of ParamMem as an effective component for enhancing language agents.

Tianjun Yao, Yongqiang Chen, Yujia Zheng +3

Mohamed bin Zayed University of Artificial Intelligence · Carnegie Mellon University · Georgia Institute of Technology

Self-Reflection in Language Model AgentsMemory-Augmented Language Model AgentsLLM Self-RefinementMemory-Augmented Language ModelsSelf-Improving Agents
AskPDF
Feb 25, 2026

Your Code Agent Can Grow Alongside You with Structured Memory

While "Intent-oriented programming" (or "Vibe Coding") redefines software engineering, existing code agents remain tethered to static code snapshots. Consequently, they struggle to model the critical information embedded in the temporal evolution of projects, failing to leverage the "reasoning trajectories" implicit in past successful practices. This limitation results in rigid behavioral logic and a lack of autonomous adaptability, ultimately hindering their ability to tackle complex, repository-level problems. To bridge this static-dynamic mismatch, we propose MemCoder, a framework designed to enable continual human-AI co-evolution. MemCoder first structures historical human experience to distill latent intent-to-code mappings from past commits. It then employs a self-refinement mechanism driven by verification feedback to correct agent behavior in real-time. Crucially, an experience self-internalization mechanism is introduced to crystallize human-validated solutions into long-term knowledge, thereby supporting sustained evolution. Experimental results on SWE-bench Verified demonstrate that MemCoder not only achieves State-of-the-Art (SOTA) performance but also delivers a 9.4% improvement in resolved rate over the general foundation model DeepSeek-V3.2. These findings indicate that equipping agents with the capability to co-evolve with humans via project history and real-time feedback effectively unlocks the potential of general models in complex software engineering tasks.

Yi-Xuan Deng, Xiaoqin Liu, Yi Zhang +2

Tsinghua University · Intern at Proxseer Inc · Proxseer Inc.

Software Engineering AgentsMemory-Augmented AgentsSelf-Improving AgentsContinual Learning for LLM Agents
AskPDF
Jan 18, 2026

Survival is the Only Reward: Sustainable Self-Training Through Environment-Mediated Selection

Self-training systems often degenerate due to the lack of an external criterion for judging data quality, leading to reward hacking and semantic drift. This paper provides a proof-of-concept system architecture for stable self-training under sparse external feedback and bounded memory, and empirically characterises its learning dynamics and failure modes. We introduce a self-training architecture in which learning is mediated exclusively by environmental viability, rather than by reward, objective functions, or externally defined fitness criteria. Candidate behaviours are executed under real resource constraints, and only those whose environmental effects both persist and preserve the possibility of future interaction are propagated. The environment does not provide semantic feedback, dense rewards, or task-specific supervision; selection operates solely through differential survival of behaviours as world-altering events, making proxy optimisation impossible and rendering reward-hacking evolutionarily unstable. Analysis of semantic dynamics shows that improvement arises primarily through the persistence of effective and repeatable strategies under a regime of consolidation and pruning, a paradigm we refer to as negative-space learning (NSL), and that models develop meta-learning strategies (such as deliberate experimental failure in order to elicit informative error messages) without explicit instruction. This work establishes that environment-grounded selection enables sustainable open-ended self-improvement, offering a viable path toward more robust and generalisable autonomous systems without reliance on human-curated data or complex reward shaping.

Jennifer Dodgson, Alfath Daryl Alhajir, Michael Joedhitya +6

Self-TrainingSelf-Improving AgentsSelf-Evolving Agents
AskPDF
Nov 27, 2025

Co-Evolving Agents: Learning from Failures as Hard Negatives

Self-evolving agents improve their performance on long-horizon tasks by learning from their own interactions with an environment. A common approach uses the resulting failed trajectories as negatives for preference training. However, collecting an agent's own failures does not ensure that they provide informative supervision for further improvement. Obvious failures may be easy to reject without learning to identify errors in more plausible attempts. These plausible but incorrect trajectories can serve as hard negatives, helping agents learn to distinguish successful behavior from failures with less obvious errors. To enable agents to generate and learn from such informative negatives, we propose a co-evolving framework that couples a target agent's self-improvement with a failure agent trained to generate hard negatives. The failure agent learns exclusively from both agents' failed trajectories to generate higher-reward failures, which the target agent uses as negatives for preference training. As the target policy improves, its failures provide new training data for the failure agent. The updated failure agent, in turn, provides new negatives for target training, allowing negative generation to adapt alongside the target policy. Across online shopping, scientific reasoning, and interactive SQL querying, our framework improves average task reward by 5.7 over the baseline. These results demonstrate the value of learning to generate and use informative failures for agent self-improvement. We will make our code publicly available.

Yeonsung Jung, Trilok Padhi, Sina Shaham +4

KAIST · Georgia State University · Independent Researcher +2

Hard Negative MiningSelf-Improving AgentsPreference LearningSelf-Evolving Agents
AskPDF
Oct 11, 2025

SGM: A Statistical Godel Machine for Risk-Controlled Recursive Self-Modification

Recursive self-modification is increasingly central in AutoML, neural architecture search, and adaptive optimization, yet no existing framework ensures that such changes are made safely. Godel machines offer a principled safeguard by requiring formal proofs of improvement before rewriting code; however, such proofs are unattainable in stochastic, high-dimensional settings. We introduce the Statistical Godel Machine (SGM), the first statistical safety layer for recursive edits. SGM replaces proof-based requirements with statistical confidence tests (e-values, Hoeffding bounds), admitting a modification only when superiority is certified at a chosen confidence level, while allocating a global error budget to bound cumulative risk across rounds.We also propose Confirm-Triggered Harmonic Spending (CTHS), which indexes spending by confirmation events rather than rounds, concentrating the error budget on promising edits while preserving familywise validity.Experiments across supervised learning, reinforcement learning, and black-box optimization validate this role: SGM certifies genuine gains on CIFAR-100, rejects spurious improvement on ImageNet-100, and demonstrates robustness on RL and optimization benchmarks.Together, these results position SGM as foundational infrastructure for continual, risk-aware self-modification in learning systems.Code is available at: https://github.com/gravitywavelet/sgm-anon.

Xuening Wu, Shenqin Yin, Yanlan Kang +4

Fudan University · University of California, San Diego · Tongji University

Automated Machine LearningSelf-Improving AgentsSequential Hypothesis TestingRecursive Self-Improvement
AskPDF