PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents

• URL: http://arxiv.org/abs/2608.04003v1
• Date: Tue, 04 Aug 2026 17:58:05 GMT
• Title: PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents
• Abstract summary: PAST-Bench is a benchmark designed to isolate whether retained experience actually improves agents over time.<n>We report both later-task gains and whether those gains follow the intended save, retrieve, and update pathway.<n>We develop Hermes+, which extends Hermes with five targeted interventions across stages of the agent loop.
• Score: 50.9061759859778
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recursive self-improvement requires agents to turn accumulated experience into better future behavior. Personal AI agents offer a concrete setting for studying this capability because they retain preferences, task histories, tool routines, and learned skills across sessions. Yet whether retained experience actually improves them over time has not been systematically tested. We introduce PAST-Bench, a benchmark designed to isolate this question. Each agent runs through ordered sequences of fresh-session tasks under matched conditions that turn retained experience on and off. It spans 26 scenarios and 204 episodes across memory, procedural reuse, information gathering, and update. We report both later-task gains and whether those gains follow the intended save, retrieve, and update pathway. Across seven base models and four agent frameworks, improvement is real but uneven across capabilities. Agents with the same headline gain can differ markedly in whether that gain is supported by evidence of the intended pathway. Guided by these findings, we develop Hermes+, which extends Hermes with five targeted interventions across stages of the agent loop. Hermes+ raises the average gain from retained experience and provides clearer pathway evidence, with its strongest improvement on tasks requiring outdated state to be replaced, although the effect remains capability- and model-dependent. Together, PAST-Bench and Hermes+ provide an evaluation and diagnostic foundation for studying how persistent agents can progress from retaining experience to systematically improving through it. Code: https://github.com/Gen-Verse/PAST-Bench

Related papers

• S3Gym: Can LLMs Turn Self-Testing and Self-Judging into Self-Improvement? [66.17176838165994]
  Large language models (LLMs) increasingly interact with external environments and accumulate substantial behavioral experience.<n>It remains unclear whether an agent can actively test its behavior, judge the resulting experience, and use that experience to improve future decisions.<n>We introduce textbfStextsuperscript3Gym, an interactive benchmark for evaluating LLM self-improvement.
  arXiv  Detail & Related papers  (2026-08-31T17:05:41Z)
• PATH-Bench: Path-Dependent Evaluation of Lifelong Agents [27.837334107109786]
  PATH-Bench is a benchmark for path-dependent evaluation of lifelong agents.<n>We propose Selective Experience Use (SEU), an agent harness that regulates how path-accumulated experience influences each new task.
  arXiv  Detail & Related papers  (2026-08-02T11:00:04Z)
• Knowledge-Centric Self-Improvement [28.85741798328687]
  Self-improving AI systems treat the agent as the object that improves, by optimizing prompts, harnesses, or even the agent's own code.<n>This agent-centric view can make improvements expensive to maintain and difficult to transfer.<n>We study a complementary paradigm: knowledge-centric self-improvement, in which agents remain generic and disposable.
  arXiv  Detail & Related papers  (2026-07-21T21:38:39Z)
• TTHE: Test-Time Harness Evolution [50.26245555541721]
  Existing approaches optimize such harnesses before deployment, searching training or development data for a fixed agent workflow that is then frozen at test time.<n>We ask whether the harness can instead be optimized during evaluation itself, using only the unlabeled execution traces the agent produces on the test inputs.<n>During evaluation, TTHE maintains a population of candidate harnesses and refines them through an agentic proposer that reasons over their execution traces.<n>A judge then commits an improved harness from execution-derived proxy signals, and the selected program persists to govern subsequent inputs.
  arXiv  Detail & Related papers  (2026-07-09T05:53:39Z)
• EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer [25.657960361149833]
  EvoAgentBench is a benchmark for agent self-evolution via Ability-guided transfer.<n>Every test task is backed by verified training-side Ability support.
  arXiv  Detail & Related papers  (2026-07-06T15:17:09Z)
• SEAGym: An Evaluation Environment for Self-Evolving LLM Agents [30.362414820430402]
  SEAGym is an evaluation environment for measuring agent harness updates across training, validation, test, replay, and cost records.<n>It turns Harbor-compatible benchmarks into dynamic self-evolution task sources with train batches, frozen update-validation, held-out ID and OOD transfer views, replay diagnostics, and saved snapshot and metric records.
  arXiv  Detail & Related papers  (2026-06-16T05:50:55Z)
• Bayesian-Agent: Posterior-Guided Skill Evolution for LLM Agent Harnesses [30.457068495035386]
  textbfBayesianAgent treats reusable skills and SOPs as hypotheses about whether a frozen model will succeed under a particular prompt, context, and harness environment.<n>Bayesian-Agent records verified trajectory evidence, maintains a feature-conditioned categorical posterior over each skill, and maps posterior state into inspectable actions such as patch, split, compress, retire, and explore.
  arXiv  Detail & Related papers  (2026-06-06T21:40:19Z)
• Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems [60.79651380135334]
  Long-lived AI agents are increasingly deployed as persistent operational systems.<n>How long does an agent remain reliable after deployment?<n>We introduce AgingBench, a longitudinal reliability benchmark for agent lifespan engineering.
  arXiv  Detail & Related papers  (2026-05-25T19:55:12Z)
• Ask Only When Needed: Proactive Retrieval from Memory and Skills for Experience-Driven Lifelong Agents [21.029076711222917]
  ProactAgent is an experience-driven lifelong learning framework for proactive retrieval over a structured experience base.<n>ExpOnEvo organizes historical interactions into typed repositories, including factual memory, episodic memory, and behavioral skills, so that retrieval can provide both relevant evidence and actionable guidance.<n>ProactRL provides step-level supervision for retrieval decisions, encouraging retrieval only when it leads to better task outcomes or higher efficiency.
  arXiv  Detail & Related papers  (2026-04-22T13:50:55Z)
• RetroAgent: From Solving to Evolving via Retrospective Dual Intrinsic Feedback [54.39884046754265]
  RetroAgent is an online RL framework that enables agents to master complex interactive environments.<n>Our experiments show that RetroAgent achieves state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2026-03-09T16:23:33Z)
• AgentPRM: Process Reward Models for LLM Agents via Step-Wise Promise and Progress [71.02263260394261]
  Large language models (LLMs) still encounter challenges in multi-turn decision-making tasks.<n>We build process reward models (PRMs) to evaluate each decision and guide the agent's decision-making process.<n>AgentPRM captures both the interdependence between sequential decisions and their contribution to the final goal.
  arXiv  Detail & Related papers  (2025-11-11T14:57:54Z)
• EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systems [59.66823584073748]
  A fundamental limitation of current AI agents is their inability to learn complex skills on the fly at test time.<n>We present EvoTest, an evolutionary test-time learning framework that improves an agent without any fine-tuning or gradients-by evolving the entire agentic system after every episode.
  arXiv  Detail & Related papers  (2025-10-15T07:16:28Z)

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.