self-evolving agent

The papers

Most central to this idea first, not most recent.

• Independentcs.CR

  Poisoned training tasks can backdoor self-improving AI agents

  Poisoned evolution tasks can trick self-improving AI agents into building backdoors into their own skills.

  Nov 2026

• Chinese Techcs.CV

  AI agents can perform better while keeping fewer memories

  Memory curation by replaying source decisions before retention improves LLM agent success rates and reduces token use.

  The University of Hong Kong, Huazhong University of Science and Technology · Nov 2026

• Big Techcs.AI

  Models can build their own agent software, but improvements generalize unreliably

  LLMs can build capable task harnesses, but two rounds of self-revision do not reliably improve held-out performance and may trade performance for lower inference cost.

  Amazon, Technische Universität Dresden · Oct 2026

• Chinese Techcs.AI

  AI sales agents that learn from past chats lift conversions 16%

  Sales strategies modeled as testable hypotheses and refined across merchant conversations, without retraining, lifted conversion 15.8% and high-intent merchant growth 35.8% in a live test on 1688.

  South China University of Technology, Alibaba Group · Nov 2026

• Industrycs.RO

  Benchmark reveals when robotic agents truly learn and when they just retry

  Self-improving robot agents do not necessarily learn from experience; apparent gains can come from retries or luck, and learned skills do not always transfer.

  Luma AI · Oct 2026

• Chinese Techcs.MA

  Multi-agent AI systems improve by learning from execution traces

  MAS-Loop closes the feedback loop in multi-agent systems by using execution trajectories to refine and improve systems across training iterations.

  Hong Kong University of Science and Technology, Ant Group · Oct 2026

• Big Techcs.AI

  Self-evolving agents trade plasticity for stability in continual learning streams.

  Self-evolving LLM agents trade plasticity for stability: self-modification interferes across tasks, and memory retrieval causes collisions, so they are not complete continual learners.

  Mila – Quebec AI Institute, Université de Montréal · Oct 2026

• Independentcs.AI

  AI models improve at predicting physics when objects are removed

  VLMs trained on self-proposed, simulator-verified counterfactual physics tasks improve causal physical reasoning and transfer zero-shot to unseen dynamics.

  Oct 2026

• Academiccs.AI

  Coding up definite operations improves AI agent consistency

  Compiling well-specified skill operations into deterministic programs reduces execution uncertainty and improves AI agent task performance.

  East China Normal University, China University of Geoscience · Oct 2026

• Chinese Techcs.AI

  AI agents improve when training tasks adapt to their weaknesses

  EvoMCP improves AI agents by co-evolving training tasks and demonstrations based on student feedback.

  Shanghai Jiao Tong University, Ant Group · Nov 2026

• Big Techcs.LG

  Agents improve by distilling hindsight from their own interactions

  Retrospective Self-Distillation lets agents evolve from their own trajectories using hindsight and cross-task precedents without external supervision.

  Meta AI · Oct 2026

• Big Techcs.CL

  Coding agents improve more when their model and harness co-evolve

  Alternating self-proposed harness revisions with policy reinforcement learning improves coding-agent resolve rates beyond optimizing either component alone.

  Microsoft Research, University of Virginia · Oct 2026

• Research Labcs.AI

  Execution traces enable precise localized repair of agent memory architectures

  Localized code repair using raw traces and memory state improves agent memory architectures more than coarse feedback.

  State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences · Oct 2026

• Chinese Techcs.LG

  Guided self-exploration lets language models teach themselves better reasoning

  SEPO uses labels to guide LLM reasoning exploration, improving self-evolution efficiency and final performance.

  Renmin University of China, Peking University · Oct 2026

• Chinese Techcs.AI

  Constructing testable, stationary rewards for open-ended tool agents

  Validated, evidence-producing rewards held fixed during optimization improve the consistency of open-ended tool agents’ outputs on two office-work benchmarks.

  Huawei · Oct 2026

• Top Universitycs.CL

  Cross-fitted feedback curbs shared errors in self-evolving search agents

  Co-cheating in self-evolving search agents is mitigated by cross-fitting source documents to break feedback loops.

  Rutgers University, University of California, San Diego · Oct 2026

• Big Techcs.AI

  Agent evolvers can learn their own search strategies across tasks

  A learned campaign policy lets agent evolvers organize their own search and match or exceed hand-designed loops while transferring across tasks.

  AWS AI Labs, Georgia Institute of Technology · Oct 2026

• Independentcs.AI

  A closed-loop system lets harness agents improve without external teachers

  ClawLoop is a teacher-free closed-loop framework that lets harness-based LLM agents iteratively improve by synthesizing their own training tasks.

  Oct 2026

• AI Startupcs.AI

  Self-evolving allocation agent cuts evaluation costs in program evolution

  EvoAlloc learns from search experience to dynamically allocate evaluation resources, reducing required evaluations by 59-82% and improving final performance.

  King Abdullah University of Science and Technology (KAUST), Sakana AI · Oct 2026

• Big Techcs.LG

  AI agents learn phone app navigation on the fly without offline training

  MAPLE uses cross-episodic playbook sharing to guide mobile agents, achieving state-of-the-art task success without training.

  Google, Washington University in St. Louis · Nov 2026

• Big Techcs.LG

  Jointly co-compiling draft and target models boosts speculative decoding speed

  Conjugate Compilation jointly optimizes draft and target models for speculative decoding, achieving up to 1.59x speedup on single GPUs.

  University of California, San Diego, Massachusetts Institute of Technology · Oct 2026

• Big Techcs.LG

  Language model agents evolve data pipelines to boost tabular foundation models

  TabFM-Auto pairs a frozen tabular model with an LLM agent that iteratively refines the data pipeline, beating all competitors on 51 benchmarks.

  Google Research, University of Southern California · Oct 2026

• Chinese Techcs.CV

  Self-evolving AI agent generates cinematic multi-shot videos by iteratively directing and critiquing itself

  A unified Director Critic vision-language policy trained with alternating reinforcement learning improves cinematic multi-shot video generation.

  Beijing University of Posts and Telecommunications, Peking University · Oct 2026

• Research Labcs.CR

  Self-updating agents can improve performance while quietly weakening safety

  Self-evolving LLM agents can become unsafe through their own harmless updates, even without adversarial attacks.

  University of Virginia, ELLIS Institute Tübingen · Oct 2026

• Big Techcs.CL

  Adaptive simulated training makes web agents more resistant to prompt injection

  AdvSim2Real co-evolves tasks, attacks, and a web agent in a simulated world, raising task completion under unseen attacks by 33.6%.

  Mohamed bin Zayed University of Artificial Intelligence, Amazon · Oct 2026

• Big Techcs.CV

  Evolving physics captions improve video models’ physical plausibility

  Using a self-evolving language framework improves video world model physical plausibility without extra visual signals.

  NVIDIA, University of Oxford · Oct 2026

• Big Techcs.AI

  Verified self-improvement makes agent harness optimization more reliable

  VERSE improves harness optimizers by letting them test edits and evolve their own verification tools.

  MIT, Amazon · Oct 2026

• Top Universitycs.LGcode

  Evidence-driven harness edits improve frozen GUI agents across benchmarks

  GUI-HARVEST automatically improves GUI agent harnesses by diagnosing failures from repeated executions and generating reusable edits, boosting performance across models.

  The Chinese University of Hong Kong, Shenzhen, Tianjin University · Oct 2026

• Big Techcs.AI

  Self-evolving visual agents learn tool use from unlabeled images using margin reward.

  A margin reward based on the solver's performance difference with and without tools enables visual agents to learn tool invocation from unlabeled images.

  UCAS, Microsoft Research Asia · Oct 2026

• Top Universitycs.AIcode

  EvoSteer Repairs Multi-Agent Workflows While They Are Still Running

  EvoSteer outperforms baselines across twelve datasets with online self-evolving graph orchestration using AnchorTB and validated skill admission.

  The Chinese University of Hong Kong, Shenzhen, Dalian University of Technology · Oct 2026

• Big Techcs.LG

  Forecast errors can improve event-aware time series predictions

  SEER dynamically refines event retrieval and causal knowledge using prediction errors to improve time series forecasting.

  Google Cloud AI Research, University of Virginia · Oct 2026

• Big Techcs.CL

  Adaptive agent teams improve consistency in long-form subtitle translation

  SMART self-evolving multi-agent system improves long-form subtitle translation across 15 locales, reducing penalty by 6.9%.

  University of Illinois Urbana-Champaign, Amazon · Oct 2026

• Big Techcs.AI

  Evolving environment beliefs helps LLM agents adapt without retraining

  EvoBelief improves frozen LLM agents by validating, reusing, and revising falsifiable environment rules, including after hidden dynamics changes and during scientific discovery.

  Tulane University, Amazon · Oct 2026

• Big Techcs.AI

  Multi-objective evolution yields stronger, more transferable agent harnesses

  Harmony Harness jointly optimizes functionality and safety in LLM agent harnesses, outperforming single-objective baselines by over 10%.

  Google Cloud AI Research, Google DeepMind · Oct 2026

• Big Techcs.AI

  Agentic data synthesis improves by evolving the generation space based on learner feedback

  EvoSyn evolves a revisable synthesis space using learner feedback, reallocating and expanding generation controls, and improves agent task success by 13.3 points.

  University of Southern California, AWS AI · Oct 2026

• Big Techcs.CL

  Jointly evolving algorithmic plans and code yields fast hybrid optimizers

  HeurEvo jointly evolves algorithmic plans, code, and reusable components to design hybrid solver-augmented heuristics that perform well under tight runtime budgets.

  Purdue University, Microsoft Research · Oct 2026

• Chinese Techcs.LG

  Refining rejected candidates makes AI system optimization more efficient

  Mara Chain retains and refines rejected AI system candidates instead of discarding them, improving performance with fewer evaluations.

  Ant Group, Beijing Intelligent Game and Decision Lab · Sept 2026

• Academiccs.CL

  Coding agents can improve by evolving their own instruction rules

  RuleEvolve self-evolves coding rules by iterative mutation and selection, beating hand-written rules and prompt optimization on correctness, length, and cost.

  Duke University · Oct 2026

• Big Techcs.AI

  Agent memory improves when evolution targets capabilities separately

  Guiding agent-memory evolution by individual capability dimensions rather than overall performance uncovers and preserves improvements, yielding gains on million-token histories.

  University of Science and Technology of China, Microsoft · Oct 2026

• Top Universitycs.CL

  Self-programmed memory adapts retrieval depth to each agent query

  MemCodex improves task success by 10.1% with 3.4x fewer tokens and 2.1x faster inference.

  Université de Montréal, Mila – Quebec AI Institute · Oct 2026

• Big Techcs.AIcode

  Adaptively choosing how to revise LLM skills outperforms fixed revision strategies

  StraTune adaptively selects revision operators to improve self-evolving LLM skills, outperforming fixed operators across tasks.

  The University of Texas at Austin, Amazon · Sept 2026

• Big Techcs.AIcode

  Research-guided software evolution improves fixed-model agents on two benchmarks

  ScholarEvolve automatically uses state-of-the-art research and topic modeling to guide agent harness evolution, improving task completion on AppWorld and Tau2-Bench.

  University of California, Santa Barbara, Microsoft · Oct 2026

• Chinese Techcs.AI

  Statistical gates help self-editing agents avoid lasting performance regressions

  SAGE uses per-item paired comparisons and one-sided paired tests to accept skill edits only when wins are statistically reliable, reducing regressions in self-evolving LLM agents.

  Peking University, University of Oklahoma · Sept 2026

• Independentcs.AI

  Whole-system training improves general-purpose agents’ use of external tools

  WEFT scales tool-use post-training for general-purpose agents by coupling whole-system interaction construction, execution-driven self-evolution, and stable post-training.

  Sept 2026

• Chinese Techcs.AI

  Training on self-generated algorithms helps design agents escape performance plateaus

  Population-Curated Policy Optimization enables self-evolving LLM agents to escape in-context evolutionary stagnation by learning from self-generated algorithms.

  Nanjing University, Huawei Noah’s Ark Lab · Oct 2026

• Chinese Techcs.LG

  Averaging rollout contexts improves label-free language model training

  Self-rewarding LLMs become more reliable when reward estimates are marginalized over random group contexts, improving zero-label self-evolution.

  Shanghai Jiao Tong University, Tencent · Sept 2026

• Big Techcs.AI

  Adaptive judges help embodied agents improve as failure patterns change

  VeriFine co-evolves policy, curriculum, and judge to scale verification and enable continuous self-improvement in embodied reasoning.

  NVIDIA, UCLA · Oct 2026

• Academiccs.AI

  Local clinical agents learn to consult remote models without identifying patients

  PrivMeSA reduces patient re-identification risk to 0% and improves task accuracy by up to 15.8 percentage points.

  University of Central Florida, Université de Montréal · Oct 2026

• Big Techcs.RO

  Humanoid tracker expands executable manipulation skills from sparse human demonstrations

  InterMimicGen self-evolves motion imitation data and policies to scale humanoid loco-manipulation from sparse human demonstrations.

  University of Illinois Urbana-Champaign, NVIDIA · Oct 2026

• Top University

  Self-improving agent improves automated optical lens design across diverse tasks

  LensDesigner autonomously designs optical lenses by retrieving seeds from a library and self-evolving through progressively harder tasks.

  INSAIT, Sofia University “St. Kliment Ohridski”, Zhejiang University · Sept 2026

• Independentcs.AI

  Safe agent self-modification requires verifiable, expressive recovery mechanisms

  Reliable LLM agent self-evolution requires co-designing verification, grounding, witness semantics, and recovery expressivity.

  Independent Researcher · Sept 2026

• Academiccs.AIcode

  Memory lets traffic-simulation agents build reusable skills without retraining

  SimSkill, a self-evolving LLM agent with episodic, procedural, and semantic memory, improves verified success on SUMO benchmarks by up to 25 percentage points.

  Jilin University · Sept 2026

• Independentcs.CL

  LLM agents do not reliably improve from their own experience

  LLMs can self-improve via self-testing and self-judging, but efficacy depends on pathway and task.

  Sept 2026

• Industrycs.CL

  Self-evolving loop synthesizes high-quality multimodal training data

  VISA uses a self-evolving loop to synthesize multimodal instruction data, improving quality via verification feedback and target model errors.

  vivo AI Lab · Aug 2026

• Independentcs.AI

  Self-improving AI agents outperform fixed ones on open-ended tasks

  Recursive self-improvement helps frontier agents most when they build their own solutions and receive fine-grained feedback.

  Nov 2026

• Independentcs.CL

  AI models share core values but differ on when to stop self-editing

  Whether a language model stops self-modifying depends more on its identity than on what it is told to optimize.

  Nov 2026

• Academiccs.LG

  Agent improves by distilling its own verified experiences during deployment

  Deployed LLM agents can improve by self-distilling their own verified execution trajectories into persistent weight updates without a separate training phase.

  University of New South Wales (UNSW Sydney) · Oct 2026

• Chinese Techcs.AI

  Diagnosis-first framework boosts solver performance across numerical domains

  ADSD links numerical diagnosis to reusable solver skill discovery, improving solver accuracy, robustness, and efficiency across four numerical domains.

  University of California, Berkeley, Alibaba Group · Oct 2026

• Chinese Techcs.AI

  Safety systems that learn from runtime experience protect LLM agents

  SafeCoEvo co-evolves an external safety harness and guard at test time, cutting unsafe outcomes by 10.05% and raising task success by 12.15% over the strongest baseline.

  East China Normal University, Shanghai Innovation Institute · Sept 2026

• Chinese Techcs.AI

  Delayed recommendation feedback can guide useful evolution of agent memory

  Delayed recommendation feedback evolves agent memory via three-stage credit assignment, achieving +7.75% MEG.

  Harbin Institute of Technology, Alibaba Group · Sept 2026

• Chinese Techcs.CV

  AI video generation gains emotional fidelity when rewards trust only reliable cues

  EmoReward adaptively calibrates reward weights in emotional video generation using reliability learned from previous training run traces.

  The Hong Kong University of Science and Technology, Tencent Hunyuan · Nov 2026

• Big Techcs.SE

  Coding agents improve software speed by revising their own setup

  Agents that improve their own code permission harnesses during a task achieve higher performance speedups than offline optimization.

  Nanjing University, Zhejiang University · Nov 2026

• Top Universitycs.AI

  Coding agents working in teams can outperform expert-written AI chip code

  Evolving Agent Teams surpasses the expert-written FlashInfer MLA decode kernel, achieving 1.21x mean speedup.

  University of California, Berkeley · Oct 2026

• Chinese Techcs.AI

  Self-improving AI agents often lose their gains on harder, different tasks

  Current self-improving agents' gains are inconsistent across generalization levels.

  Shanghai Jiao Tong University, Theseus Labs · Nov 2026

• Big Techcs.AI

  AI agents make corrections nearly twice as reliable by remembering mistakes

  Self-Directed Repair makes agents learn from mistakes by storing lessons in lightweight LoRA adapters, reducing repeated errors.

  Princeton University, Meta FAIR · Oct 2026

• Independentcs.AI

  Discarding stored abstractions improves how LLM agents learn from experience

  Persisting inferred agent memories can reproduce errors, while rebuilding temporary guidance from stored trajectories improves held-out success and reduces LLM-call cost.

  Oct 2026

• Big Techcs.SE

  Terminal agents can build adaptive environments for their own training

  AutoEnvScaling uses terminal agents to automate RL environment generation, improving performance and enabling recursive self-improvement.

  Northeastern University, Microsoft · Oct 2026

• Academiccs.AI

  Agents Repair Executable Skill Packages Better With Code-Structure Guidance

  SkillScriptBench shows AST-guided, scope-limited editing substantially improves reliable repair of executable agent skills while coordinating scripts and documentation.

  The Hong Kong University of Science and Technology, Harbin Institute of Technology, Shenzhen, China · Oct 2026

• Top Universitycs.AI

  Agents improve long-horizon performance by evolving how they manage context

  Evolving context management policies from long-horizon trajectories improves long-horizon agent performance more effectively than evolving skills or using fixed context strategies.

  Fudan University, Shanghai Key Laboratory of Data Science · Sept 2026

• Chinese Techcs.AI

  RefCon improves agent memory by refining and contrasting past experience

  RefCon combines self-refinement and self-contrast to extract higher-quality memories without gold labels.

  Huawei Technologies, Co., Ltd. · Oct 2026

• Academiccs.AI

  Self-modifying coding agents evolve learned optimizers from simple templates

  HADA automates MetaBBO algorithm design using recursive self-improvement, outperforming human-designed baselines significantly.

  South China University of Technology, South China Normal University · Sept 2026

• Chinese Techcs.LG

  Adoption-aware memory learning boosts LLM agent NPU kernel synthesis

  SAGE achieves 95.5% execution rate on NPU kernel synthesis by tracking adoption and consolidating reusable rules.

  Fudan University, University of Science and Technology of China · Sept 2026

• Independentcs.CL

  Models struggle to improve themselves from vague capability goals

  Self-evolution from vague goals leads to mismatched training data and non-transferable improvements.

  Sept 2026

• Big Techcs.AI

  Memory guidelines help language-model agents succeed consistently across repeated runs

  LLM agents averaging 77% per-run success on AppWorld pass all five runs on only 53% of tasks; self-evolving consistency guidelines raise this by 16 points same-task and 13 similar-task.

  IBM Software Innovation Lab, IBM Research · Sept 2026

• Academiccs.AI

  Self-improving agent learns video-to-action control from environment interaction alone

  Mirror learns video-to-action control from environment interaction alone, following visual plans without rewards or expert action labels and outperforming prior learning-from-observation methods.

  Massachusetts Institute of Technology · Oct 2026

• Big Techcs.AI

  Terminal-only control improves digital agents across web, desktop and CAD

  A terminal-only harness using bash commands outperforms GUI harnesses across web, desktop, and CAD tasks.

  Microsoft Research, National University of Singapore · Oct 2026

• Chinese Techcs.CV

  Video agents improve their own harnesses using unlabeled video

  VideoEvoHarness evolves video agent harnesses from unlabeled videos without annotations or parameter updates, improving accuracy.

  Institute of Automation, Chinese Academy of Sciences, University of Chinese Academy of Sciences · Oct 2026

• Chinese Techcs.AI

  AI web agents learn long tasks by reusing skills salvaged from failures

  SCOPE achieves 3.9% to 18.6% relative gains on web tasks by progressively discovering and composing local skills.

  Renmin University of China, Huawei Technologies · Nov 2026

• Top Universitycs.AI

  AI agents can rewrite their improvement routines, outperforming human-written instructions

  Skill∞ recursively improves both task skills and the meta-skill used to optimize them.

  National University of Singapore, City University of Hong Kong · Oct 2026

• Academiccs.AI

  GUI agents improve more when experience is routed by component

  GUI agent experience should be split into locators, procedures, state facts, and lessons, with locators and lessons in weights and procedures and state facts in context.

  South Dakota State University · Oct 2026

• Big Techcs.AI

  A verifiable runtime for agents that extend themselves with safe, reversible tools.

  A Lean 4 agent runtime uses reversibility to decide human oversight, letting agents extend their own tools and act autonomously.

  Amazon · Oct 2026

• Big Techcs.LG

  Prompt evolution lets frozen LLMs learn new forecasting regimes without retraining

  Evolving a short guidance prompt lets a frozen LLM outperform time-series foundation models on held-out multimodal forecasting without retraining.

  Amazon Web Services, University of California · Oct 2026

• Chinese Techcs.AI

  MetaExp trains a transferable experience-learning capability in language model agents

  MetaExp trains language models to learn and reuse experience across tasks using verifiable cross-task families.

  University of Macau, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences · Oct 2026

• Chinese Techcs.CV

  Evolved agent harnesses cut long-video analysis costs while improving accuracy

  VidHarness automatically evolves video agent harnesses, achieving state-of-the-art long video QA with fewer frames.

  Baidu, Inc. · Oct 2026

• Industrycs.AI

  Research agent improves itself through seven successive code rewrites

  An AI research agent can improve its own research efficiency by rewriting its own harness, and those gains transfer to held-out tasks.

  Weco AI · Sept 2026

• Big Techcs.AIcode

  Structured reasoning framework improves LLM game-playing efficiency and payoff

  Structuring LLM strategic reasoning as anchor, adapt, recalibrate improves payoff and cuts token use in imperfect-information games.

  The Hong Kong University of Science and Technology (Guangzhou), Johns Hopkins University · Sept 2026

• Chinese Techcs.AI

  Contrasting multiple trajectories sharpens skill learning for LLMs

  SkillCome improves skill evolution for language models by contrasting multiple successful and failed trajectories and using dual memory to derive reusable skills.

  Fudan University, Ant Group · Sept 2026

• Chinese Techcs.AI

  Branching search improves how language agents evolve reusable skill libraries

  SkillVine uses branching exploration to evolve LLM agent skill libraries, outperforming linear evolution on nine of ten benchmark-model combinations.

  The Chinese University of Hong Kong, Noah’s Ark Lab, Huawei Technologies · Sept 2026

• Research Labcs.CV

  Visual agents improve by jointly evolving reusable skills and practice data

  V-Gym co-evolves procedural skills and multimodal practice data from agent execution trajectories for continual self-improvement.

  State Key Laboratory of AI Safety, Institute of Computing TechnologyChinese Academy of Sciences · Sept 2026

• Big Techcs.AI

  AI agents improve when their surrounding software is automatically redesigned

  HarnessFlow achieves the highest mean test scores on six benchmarks by using MCTS-guided search to optimize LLM agent harnesses.

  South China University of Technology, Fudan University · Nov 2026

• Industrycs.LGcode

  Reusing agent trajectories cuts token use on repeated tool tasks

  TRIAGE reduces token consumption in LLM agents by reusing historical execution trajectories through a three-level routing framework.

  China Telecom Cloud · Sept 2026

• Independentcs.AI

  Five-level ladder maps reasoning models beyond direct human oversight

  Autonomous rewards and self-generated experience enable large reasoning models to improve beyond human supervision.

  Sept 2026