Deterministic Replay

Follow this topic Ask about this topic

Momentum

4 papers in the last four weeks, up 33% on the four weeks before. 0.0% of all new papers.

Jul 6Week of Sep 21

Latest papers 31

Sep 27, 2026cs.AI

EverMine: Dissecting the Self-Evolution of Research Capabilities in Long-Horizon Alpha Research

Self-evolving agents aim to turn research feedback into reusable skills, tools, and research rules. Whether these accumulated capabilities continue to improve later research requires controlled evaluation. Long-horizon alpha discovery provides a state-dependent setting: once a new factor enters the portfolio, the predictive information already covered changes, so the value of the same candidate or experience may change over time. We introduce EverMine, an empirical framework for studying self-evolving research capabilities in long-horizon alpha discovery. EverMine decomposes the research state into history (Hist), the current factor portfolio (Frontier), and reusable capabilities (Cap). Under matched resource limits, we compare complete runs with fixed or evolving Cap, and replace Cap while holding Hist and Frontier fixed to estimate the conditional value of accumulated capabilities. We also combine full trajectories with historical-state replay to examine how experience-based decisions affect candidate selection and portfolio outcomes. Across 18 long-horizon trajectories, end-to-end comparisons show no consistent gain from Cap evolution. Across 48 continuation branches from shared Hist and Frontier states, accumulated Cap also does not consistently outperform the initial Cap. Parameter tuning of existing factor structures can still improve the portfolio. In an exploratory replay of two screening batches from one Evolving trajectory, some screened-out candidates have positive marginal value at the original state, yet submitting all screened-out candidates sequentially slightly lowers final portfolio IC in both batches. These results show that candidate value depends on the evolving portfolio and submission order, and motivate evaluating self-evolving research capabilities through end-to-end outcomes, conditional capability value, and the consequences of experience-based decisions.

Siyuan Li, Jiangfeng Zhang, Rui Yao +3

The Hong Kong University of Science and Technology (Guangzhou) · The Hong Kong Polytechnic University · Singapore Management University

Self-Evolution Portfolio Construction Long-Horizon Task Planning Deterministic Replay

Sep 24, 2026cs.SE

Automatic Harness Evolution for Hardware Design Verification: Can LLMs Consolidate Gains Across Discovered Harnesses?

Agent behavior depends on the harness surrounding a language model, but it remains unclear whether language models can reliably improve such harnesses for hardware-design tasks. We study automatic harness evolution around a fixed subject model on 12 proprietary design-verification root-cause localization tasks. Across five trials per task, automatically evolved harnesses increased completed attempts by 71-76% and any-hit task coverage by 80-100%, while total correct attempts improved by only 18-24%. The strongest success reproducible at least twice result improved by one task, and later candidates exchanged gains across tasks rather than preserving them. An auxiliary candidate improved on a four-task validation set excluded from search but tied its baseline on a subsequent 12-task replay containing both search and validation tasks, so the selected gain did not persist across the full pool. Across the tested lineage, useful search, evidence, and finalization behaviors appeared in different candidates but did not consistently consolidate into a single harness that dominated across tasks and metrics. In a separate CVDP cross-benchmark case study, an automatically evolved defined-width repair harness produced 35.6% more functional passes than its 142-task reference baseline; the final functional verifier scored completed outputs but was not shown to the subject agent during repair. These results support archive-aware selection when evolution yields complementary specializations without consistent consolidation.

Kidus Seyoum, Ajay Mittur

NVIDIA

Automated Heuristic Design Deterministic Replay Root Cause Analysis

Sep 22, 2026cs.LG

Beyond Class Marginals: Bounding Rehearsal Gaps without Freezing Class Co-occurrence

Class-balanced replay controls class frequency but does not determine the interval between successive replay appearances of a class. We study this interval, the rehearsal gap, separately from the class marginal and class co-occurrence, and introduce randomised-pass replay (RPR), which visits each resident class once per shuffled pass. For a fixed set of C resident classes and replay batch size b less than or equal to C, RPR preserves the balanced time-averaged class marginal and bounds every gap by 2*ceil(C/b)-1; a churn-conditional bound applies while the resident set changes. The scheduler uses no future class information and adds no replay examples or forward passes. In a linear-head ER-ACE diagnostic, joint absence from the incoming and replay batches produces a one-sided classifier-bias gradient. Longer absence episodes are associated with larger negative bias displacement, and removing the incoming-loss mask attenuates the scheduling effect. In the primary ER-ACE experiments, RPR improves final average accuracy by 0.72-1.67 percentage points relative to independent class-balanced retrieval under reservoir storage, with positive effects also observed under balanced storage. Pretrained ViTs show positive effects on the tested LT10 streams with small replay batches, while matched larger-batch controls show no material effect. Fixed-cycle and reused-pass controls change more than one temporal statistic, so the experiments do not isolate rehearsal-gap length from all other forms of temporal dependence. The accuracy effects depend on the learner and operating regime.

Congren Dai, Nat Roongjirarat, Fei Ye

Imperial College London · King’s College London · University of Electronic Science and Technology of China

Layer-Selective Rehearsal Strategy Deterministic Replay Precision Recall Bounded-Memory Scale Gap Token Co-Occurrence Graphs

Sep 8, 2026cs.AI

EvolveScaler: Synthesizing Information-Evolution Contexts via Executable State Machines and Natural-Language Rendering

In persistent interactions, long contexts may encode an evolving process rather than a fixed record: later events can revise or revoke earlier information, changing what remains valid and what conclusions follow. We call this setting information evolution (IE). Solving IE requires identifying valid records, applying updates in order, and reconstructing the query-relevant state from the event history. Existing text-first synthesis pipelines make such data difficult to verify because state transitions and answer logic remain implicit. We introduce EvolveScaler, a code-driven framework that defines information evolution before rendering it as natural language. Human-authored operational specifications define state transitions, record validity, difficulty controls, and executable answer logic; a strong LLM then synthesizes a self-contained simulator from each specification. Executing validated simulators produces natural-language multi-turn event histories, while deterministic replay computes reference answers and atomic checklists. We instantiate EvolveScaler with 117 task prototypes and 159 final-question operators across five difficulty levels spanning approximately 7 to 1,200 events per instance, yielding about 35,100 training examples and 585 validated evaluation instances. On the very_long tier, the strongest model reaches 59.3% avg@5, while six models score below 10%. Training an internal A3B model on 6,000 EvolveScaler examples improves performance over its base checkpoint on all eight independently constructed out-of-distribution benchmarks, with a 5.25-point average gain. These results show that code-driven IE synthesis provides both challenging evaluation and transferable training supervision.

Ziliang Zhao, Zenan Xu, Shuting Wang +6

Gaoling School of Artificial Intelligence, Renmin University of China · LLM Department, Hunyuan Team, Tencent · The Chinese University of Hong Kong

Alphaevolve Automata Deterministic Replay

Aug 8, 2026cs.LG

The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World

LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents. Yet agentic routers are evaluated like single-turn routers: by replaying logged trajectories and substituting another model's recorded outputs, assuming the rest of the trajectory is unaffected. We test this assumption with branching rollouts: we fork live SWE-bench agent trajectories at controlled points, rebuild the environment, continue each fork with a different model, and compare against same-model control forks that isolate sampling and replay noise. Across six paired runs (~900 rollouts), swaps exceed their matched control floors by +0.25 to +0.66 normalized edit distance (multiplicity-corrected CIs exclude zero), rewriting 61-94% of post-fork actions; 74-77% of early swaps diverge at the first post-fork action, versus 6-35% of controls, leaving only 3% of replayed states valid. Divergence decreases with fork depth in both directions. All five outcome flips we observe occur in swap arms, upgrades rescuing unsolved instances and a downgrade losing the sole solve, and zero occur across 359 control forks. Scoring these same swaps with a log-stitching replay evaluator, replay mispredicts every success-relevant outcome call and predicts patches with 0.00-0.11 similarity to reality. Auditing the noise floor, temperature-0 "determinism" is configuration-dependent: FP8-served controls diverge on over 90% of forks while AWQ-served ones remain near-identical; and under tight budgets the stronger model more often exhausts its steps without submitting. Replay-based benchmarks score the wrong world for agentic routing; we release our harness and all trajectories.

Ashritha Gonuguntla

Carnegie Mellon University

Deterministic Replay Replay

Aug 5, 2026eess.SY

Toward Integrating Adaptive Experience Replay and Online Uncertainty Estimation in Safe Actor-Critic Optimal Control

Safe actor-critic control often treats barrier filtering, uncertainty estimation, and experience replay as separate modules, even though each changes the data used for learning and control. We develop an integrated architecture in which the uncertainty estimate updates the obstacle geometry used by a control barrier function, filter interventions and estimation residuals determine replay priority, and the critic learns from the executed rather than nominal action. We instantiate the architecture on a two-dimensional robot-navigation task with corrupted obstacle measurements and compare six component-matched configurations under common training budgets, random seeds, sensor streams, exploration, and disturbances. Evaluation includes a moderate post-training test, an eleven-level perception-noise sweep, and an exploratory extreme-stress test at multiplier

6.0

. In the extreme test, the integrated configuration recorded no contacts and reached the goal in all five evaluation seeds. Its mean cost was

7.63\pm0.44

and its obstacle-belief root-mean-square error was

3.52\pm0.55

cm. The uncertainty-estimation ablation also recorded no contacts but reached the goal in four of five seeds, with mean cost

8.96\pm2.08

and belief error

11.08\pm1.23

cm. A finite-training bound clarifies replay exposure, and a robust barrier condition states the required estimation-error and feasibility assumptions. The results support coupling estimation, safety filtering, and replay on this benchmark; broader safety and convergence claims require further study.

Mahshad Rastegarmoghaddam, Davoud Nikkhouy, Shima Samadzadeh

Department of Mechanical Engineering, Politecnico di Milano, Milan, Italy

Control Barrier Functions Soft Actor-Critic Prioritized Experience Replay Obstacle Avoidance Deterministic Replay Robot Navigation

Aug 3, 2026cs.AI

Beyond Single-Use Tokens: Durable Authorization State for Replay-Resistant LLM Agent Actions

Tool-using large language model agents frequently replan, retry failed operations, delegate tasks, and resume after crashes. These behaviors can cause one user authorization to be requested and executed multiple times under freshly issued token identifiers, even when each individual token is single-use. We call this failure semantic replay: exceeding the execution budget of a token-independent authorization instance rather than merely reusing an old token identifier. We show that identifier-local token consumption cannot prevent fresh reissuance unless the issuer retains monotonic durable state over the authorized action, confirmation event, and remaining execution budget. We introduce CapLease, an authorization-consumption layer that follows proposal- and authority-level defenses, binds an authenticated user confirmation to a canonical action, and enforces transactional Issue-Prepare-Commit transitions. Across LLM-agent replanning, retry, delegation, concurrency, confirmation-replay, and crash-recovery scenarios, identifier-local tokens permit fresh semantic reissuance, whereas CapLease and an equally stateful Server Ledger prevent duplicate admission and, with an idempotent sink, duplicate external effects. Our results identify durable authorization state, rather than token representation alone, as the systems requirement for replay-resistant agent execution.

Jinghan Xu, Longze Fan, Zeyuan Wang +2

1Nankai University, China · 2China University of Petroleum, China · 3Sun Yat-sen University, China +1

Deterministic Replay Commits

Jul 31, 2026cs.CV

TOOD: Task-Aware Out-of-Distribution Score Calibration for Continual Learners

The primary challenge of continual learning (CL) systems is to learn new tasks while remaining performant on previously learned tasks. A similarly important though less well-studied aspect of CL systems is their ability to distinguish inputs that are unlikely to come from within the set of tasks the system has already encountered, often called out-of-distribution (OOD) detection. This paper presents several findings related to the dynamics of OOD detection in CL systems, causes of performance degradation over time which we call OOD forgetting (OODF), and proposed mitigation strategies for this degradation. Chiefly, we find the unintuitive result that OODF is only weakly anti-correlated with classification performance on previous tasks, suggesting that the underlying mechanisms producing OODF are distinct. Moreover, this effect is observed for both energy-based and feature-based OOD detection methods. Energy-based detectors suffer a drop in logit scale as additional tasks are learned, which we term the Confidence Gap, while feature-based detectors also degrade under a complementary effect we call Manifold Crowding. Motivated by these observations, we propose TOOD, a training-free post-hoc method that decomposes logits into per-task energy scores and re-calibrates them using replay-buffer statistics. Experiments on CIFAR-10, CIFAR-100, and a 100-task ImageNet-1K stream show that TOOD improves OOD detection performance over uncalibrated energy in most settings and ranks first or second in nine of ten CIFAR configurations, with the largest gains when the confidence gap is most severe. These results suggest that a substantial portion of OOD deterioration in continual learning arises from score miscalibration rather than from a complete loss of discriminative structure.

Mostafa ElAraby, Samer B. Nashed, Liam Paull

DIRO, Mila - Quebec AI Institute · Universit´e de Montr´eal · Montreal, Canada +1

Out-Of-Distribution Detection Out-Of-Distribution Continual Learning Catastrophic Forgetting Deterministic Replay

Jul 17, 2026cs.SE

Agentic Synthesis against Counterexample-Supplemented Sketches

Coding agents can fix a failing example without preserving the domain rule that made it fail. We present agentic synthesis against counterexample-supplemented sketches, a repository-native method for systems whose policy is discovered during implementation. A human starts with a partial sketch, and a coding agent compiles a replaceable projection. When simulation exposes missing or mistaken policy, an operator approves the corrected behavior and the minimum general rule the case authorizes. Every Developer call names its change authority and the rules, holes, anchors, and approved behavior that must survive. Conflict or ambiguous permission leaves the files unchanged and produces a clarification question. A complete archive preserves provenance; a curated regression set gates distinct boundaries. Before another candidate is revealed, the active case and curated regressions must pass both deterministic approved-output comparison and a separate review against the current sketch. Periodic clean regeneration tests whether the sketch carries the learned policy. We demonstrate the method with CatSynth, a captured synthetic application. In one open-world run with GPT-5.4-mini, 8 of 14 frozen candidates became counterexamples. Under the corrected protocol, replay-all, evolved-sketch rebuild, and retained Sketch-CE each passed all 8 accepted cases. They passed 14, 17, and 16 of 21 withheld cases, respectively. Sketch review rejected premature empty-input and tag policies and restored dropped anchors; adjudicated reviewer errors did not become policy. One model and one reveal order cannot establish general correctness or superiority. On this suite, the second check exposed drift hidden by deterministic replay, and the reviewed sketch passed three more withheld cases than raw example replay.

Muness Castle, Eric Rubeck

Independent

Coding Agents Counterexample Deterministic Replay

Jun 16, 2026cs.SE

Execution-bound advisory automation for agentic AI: a reproducible AIBOM-driven CSAF-VEX framework

A protocol driven framework is presented that binds SBOM and AIBOM artefacts to deterministic environment capture and structured runtime telemetry. Exploitability is computed from declared artefacts, observed activation conditions, and enforced execution policies. CSAF VEX advisories are generated from combined static and runtime evidence, cryptographically signed, and validated through deterministic replay. Evaluation uses approximately 10000 component entries across synthetic Agentic AI workloads 50 to 5000 components, incorporating OSV, GitHub Advisory, KEV, and EPSS datasets.

Petar Radanliev, Omar Santos, Carsten Maple +1

Department of Computer Sciences, University of Oxford, Wolfson Building, Parks Rd, Oxford OX1 3QG · The Alan Turing Institute, British Library, 96 Euston Rd., London NW1 2DB · Cisco Systems, RTP, North Carolina, United States +2

Security Evaluation Runtime Enforcement Exploitation Multivariate Telemetry Data Deterministic Replay

Jun 15, 2026cs.SE

Bistable by Construction: Wall-Clock-Calibrated State Monitors Have No Moment-Detection Regime at Agent Cadence

Runtime monitors for autonomous agents commonly threshold an accumulated internal state - a behavioural baseline, a drift statistic, or, in our prior work, a modelled affective state. We previously reported a State Saturation Trap: threshold-on-state triggers over a continuous affect engine become near-constant alarms on SWE-bench debugging agents (Modgil 2026). A post-release audit found the engine received dt=0 between actions, so its exponential decay never operated: the published trap is a pure-accumulator result. We correct the record (erratum, v2) and treat the flaw as an experiment. The key variable it exposes is whether a monitor's dynamics are calibrated in sample time (per observation, as in CUSUM) or wall-clock time (half-lives in seconds, as in affect models and EMA baselines). On fixed-rate streams these coincide; on agent streams, where inter-action time varies by orders of magnitude, they do not. A pre-registered sweep over uniform intervals (dt in {0..600}s) on 20 trajectories shows the wall-clock level trigger has two regimes: at dt<=1s a constant alarm (20/20; median 18 firings); at dt>=60s silent. Every critical dt lies in (1,30]s. Real agent runs measure latency at median 1.53s (p90 2.33s); real coding cadence sits inside the trap regime, vindicating the empirical finding under a corrected mechanism. The structure is a property of the calibration class, not the engine: a minimal wall-clock accumulator over the raw error stream reproduces the same cliff, while a sample-time CUSUM over the identical stream is exactly dt-invariant (20/20). A rising-edge trigger with hysteresis fires 0-3 times per trajectory in every condition. We conclude that wall-clock-calibrated leaky-integrator monitors admit no regime in which they act as moment detectors on agent streams; transition detection escapes the trap at every cadence, but does not recover human intervention timing.

Manvendra Modgil

Modint Intelligence

Runtime Enforcement State-Tracking Autonomous Agents Deterministic Replay Decision Thresholds Agent Behavior Modeling

Jun 15, 2026cs.LG

Verified Detection and Prevention of Concurrency Anomalies in Multi-Agent Large Language Model Systems

Multi-agent LLM systems share state through memory stores, vector indices, and tool registries. We model such sharing as long-running read-generate-write operations under deterministic-generation semantics -- the regime durable-execution engines enforce by deterministic replay -- and formalize four concurrency anomalies in TLA+: stale-generation, phantom-tool, causal-cascade, and tool-effect reordering, structural analogues of classical isolation anomalies, each with a TLC counter-example. The exclusion lattice over these anomalies is trivial; the contribution is the mechanically verified realizability and strict separation of one maximal chain within it,

L_0 \subsetneq \cdots \subsetneq L_4

, to our knowledge the first machine-checked consistency hierarchy for such runtimes. A development of 274 Verus obligations (zero assume, zero admit; trust base: two structural axioms and a mutex correspondence) proves the detectors sound and complete against the specifications and each runtime its avoidance set. Three deployed Rust runtimes realize L0-L1 (pessimistic locking, serializable snapshot isolation, default-SI), each verified against stale-generation and refined to its state machine; L2-L4 are exec-mode-verified with dependency-free prevention twins (A3, A6, A2: 0/1000 versus 1000/1000), and L2 is run live across three model families (A3 prevented in all 120 retracted sessions). We reproduce a silent lost update in ByteDance's deer-flow, formalizing its fix as a verified

L_0 \to L_1

refinement, and exhibit tool-effect reordering in LangGraph's ToolNode on unmodified output, removed by an L3 commit-order sequencer. The verified detector, refinements, and realizability artifacts are the contribution; the phenomena and lattice are classical.

Sajjad Khan

independent researcher

Multi-Agent Large Language Model Systems Deterministic Replay Possibilities Lattice

Jun 10, 2026cs.AI

DFAH-Bench: Benchmarking Observable Agent Instability in Financial Decision-Making

A financial agent can repeat a decision while changing the work behind it. DFAH-Bench operationalizes the Determinism--Faithfulness Assurance Harness (DFAH), pairing decision agreement with tool-path agreement on the same qualified replays, then extends that qualification principle to evidence, authorization, execution and task outcomes. Retrospective and prospective replay analyses expose process variation behind stable decisions. Across 570 eligible prospective episodes, decision agreement is 94.2-95.1%, while agreement on ordered tools, arguments and results is 45.0-51.5%; one stratum falls one group below its prespecified coverage minimum. A separate capture diagnostic shows that systematic omissions can preserve perfect replay agreement. Using the

τ

-Knowledge banking environment, we retain 1,080 scheduled episodes and 1,033 known native outcomes across separate cohorts with open-weight and frontier generators. Missing outcomes prevented the planned tests, so comparisons are descriptive. On the primary schedule, structural checks alone yield more successes than either gate-and-recovery bundle. The typed-choice bundle has lower mean episode cost than the generative bundle on complete task pairs, but produces fewer successes under every assignment of unknown outcomes. Input limits and recovery behavior materially shape these results. Fixed-state probes reveal higher decision agreement alongside lower agreement with constructed policy labels, and separately expose sensitivity to retained generator rationale in a selected authorization case. Together, the findings connect replay observability to evidence, authorization, completion and cost: evidence sufficiency needs direct assessment alongside repeatability.

Raffi Khatchadourian

IBM Financial Services Market New York, NY, USA

Deterministic Replay Reproducibility Decisions Agreement Instability

Jun 1, 2026cs.LG

Quantifying the Energy Floor: Direct Measurement and Replay Buffer Bias in SAC-Based HVAC Control on sbsim

We quantify the energy floor -- the minimum achievable cost given action space constraints -- for Soft Actor-Critic (SAC) HVAC control on the sbsim calibrated building simulator. Through minimum-action experiments, we directly measure this floor at USD 35.51/day, dominated by continuous electrical loads (USD 35.44, 99.8%) with negligible gas consumption. The standard SAC baseline, initialized with schedule-policy replay buffer transitions, converges to USD 37.18/day, 4.7% above the floor. We identify buffer initialization as the dominant source of sub-optimality in this scenario: training from an empty buffer reduces cost to USD 35.57/day, eliminating 96% of the gap. Expanding the supply water temperature range by 10 K yields negligible additional savings (USD 0.03/day), and further expansion triggers physical constraint violations. We additionally uncover a discount factor coupling (gamma_eff = 0.891) shrinking the effective planning horizon from 8.3 h to 46 min -- a benchmark-wide issue warranting audit. Systematic ablation across planning horizon, reward weights, and observation enrichment confirms all pre-filled-buffer configurations cluster within 0.7% (USD 37.18--USD 37.42), demonstrating that equipment minimum power -- not algorithmic design -- imposes the binding constraint.

Bo Li, Chen Zhang

Shanghai Jiao Tong University College of Smart Energy Shanghai, China

Hvac Control Energy-Efficient Energy Resources Nvidia Isaac Sim Deterministic Replay

May 31, 2026cs.LG

RLVR without Ineffective Samples: Group Prioritized Off-Policy Optimization for LLM Reasoning

Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, its effectiveness is substantially hindered by the prevalence of ineffective training data: many sampled prompts yield response groups that are either entirely correct or entirely incorrect, resulting in zero-variance rewards and limited learning signals. Recent state-of-the-art methods address this issue through extensive LLM rollouts to filter ineffective samples, but at the cost of considerable computational overhead. Alternative approaches, including predictive sampling and trajectory replay, aim to improve data efficiency but often remain insufficient and may introduce additional issues such as systematic bias or suboptimal constraints. To address these limitations, we propose Group Prioritized Off-Policy Optimization (POPO), a simple yet effective framework that fully exploits effective training batches without additional rollout overhead. POPO comprises two key components: prioritized group replay and decoupled off-policy optimization. The former replaces ineffective on-policy groups with effective off-policy groups via a recency-based replay mechanism that jointly considers sample quality and the degree of off-policiness. To further mitigate the off-policy gap, POPO employs decoupled importance sampling to correct off-policy bias while maintaining stable policy updates under consistent trust-region constraints. Empirical evaluations across diverse reasoning tasks, including mathematics, planning, and visual geometry, demonstrate that POPO substantially accelerates RL finetuning and achieves strong reasoning performance with significantly fewer rollouts.

Yixiu Mao, Yun Qu, Qi Wang +2

Department of Automation, Tsinghua University

Reinforcement Learning With Verifiable Reward Better Call Group Relative Policy Optimization Off-Policy Learning Deterministic Replay

May 23, 2026cs.LG

Streaming Reinforcement Learning under Partial Observability with Real-Time Recurrent Learning

Streaming reinforcement learning has emerged as an online learning paradigm that conforms to the restrictions of natural learning agents that process data incrementally, i.e. with a batch size of 1 and no replay buffer. While streaming RL has recently been shown to scale with deep function approximation with full observability, partially observable settings have remained out of reach. Truncated backpropagation through time collapses to a one-step gradient horizon under the streaming setting, and exact real-time recurrent learning is prohibitively expensive. We close this gap using recurrent trace units, a diagonal recurrent architecture that enables exact RTRL with linear time and memory complexity in the parameter count, and show that they integrate cleanly into existing streaming algorithms across both discrete and continuous control. On a MemoryChain diagnostic with chain lengths from 2 to 128, our method sustains performance where streaming TBPTT(1) baselines using feedforward, GRU, and RTU networks collapse. On five POPGym tasks and on partially observable MuJoCo continuous control, the streaming approach is competitive with batched PPO on POPGym and recovers a substantial fraction of batched performance on masked MuJoCo, despite using no replay buffer or batched updates.

Noah Farr, Aryaman Reddi, Carlo D'Eramo +1

Technical University of Darmstadt · Zuse School ELIZA · Center for Artificial Intelligence and Data Science, University of Würzburg +1

Recurrent Neural Networks Deterministic Replay Partial Observability Mujoco

May 21, 2026cs.AI

The Log is the Agent: Event-Sourced Reactive Graphs for Auditable, Forkable Agentic Systems

Most agent frameworks are built around the language model: a conversation loop comes first, then tools, then rules, and finally a logging layer bolted on for observability, with state persisted as retrievable "memory." We describe ActiveGraph, a runtime that inverts this arrangement. The append-only event log is the source of truth; the working graph is a deterministic projection of that log; and behaviors--ordinary functions, classes, LLM-backed routines, or logic attached to typed edges--react to changes in the graph and emit new events. No component instructs another; coordination happens entirely through the shared graph. This single design decision yields three properties that retrieval-and-summarization memory systems do not provide: deterministic replay of any run from its log, cheap forking that branches a run at any event without re-executing the shared prefix, and end-to-end lineage from a high-level goal down to the individual model call that produced each artifact. We present the architecture, a determinism contract that makes replay sound, and a worked diligence example whose full causal structure is reconstructable from the log alone. We discuss--without claiming to demonstrate--why this substrate is unusually well suited to self-improving agents, and how it extends the BabyAGI lineage and prior graph-memory research.

Yohei Nakajima

Untapped Capital

Graph-Structured Memories Production Agentic Systems Self-Improving Agents Deterministic Replay Graph Representations

May 14, 2026cs.AI

Good to Go: The LOOP Skill Engine That Hits 99% Success and Slashes Token Usage by 99% via One-Shot Recording and Deterministic Replay

Deploying AI agents for repetitive periodic tasks exposes a critical tension: Large Language Models (LLMs) offer unmatched flexibility in tool orchestration, yet their inherent stochasticity causes unpredictable failures, and repeated invocations incur prohibitive token costs. We present the LOOP SKILL ENGINE, a system that achieves a combined 99% success rate and 99% token reduction for periodic agent tasks through a one-shot recording, deterministic replay paradigm. On its first run, the agent executes the task with full LLM reasoning while the system transparently intercepts and records the complete tool-call trajectory. A greedy length-descending template extraction algorithm then converts this recording into a parameterized, branch-free Loop Skill -- a deterministic execution plan that captures the task's functional intent while parameterizing time-dependent and result-dependent variables. All subsequent executions bypass the LLM entirely: the engine resolves template variables against real-time values and replays the tool sequence deterministically. We prove two theorems: (1) Replay Determinism -- the step sequence of a validated Loop Skill is invariant across all future executions; (2) Write Safety -- concurrent access to persistent configuration is serialized through reentrant locks and atomic file replacement. Across a benchmark of periodic agent tasks spanning intervals from 5 minutes to 24 hours, the Loop Skill Engine reduces monthly token consumption by 93.3%--99.98% and cuts execution latency by 8.7x while eliminating output non-determinism. A multi-layer degradation strategy guarantees that tasks never stall. We release the engine as part of the buddyMe open-source agent framework.

Xiaohua Wang, Kai Yu, XuXiao Liang +2

1Artificial Intelligence Research Center, Bengbu Medical University, Bengbu 233000, China. · 2CHARMMIRAEL Biotech Co., Ltd, Nanjing 210000, China.

Agent Loop Deterministic Replay Skills

May 11, 2026cs.LG

When Does Non-Uniform Replay Matter in Reinforcement Learning?

Modern off-policy reinforcement learning algorithms often rely on simple uniform replay sampling and it remains unclear when and why non-uniform replay improves over this strong baseline. Across diverse RL settings, we show that the effectiveness of non-uniform replay is governed by three factors: replay volume, the number of replayed transitions per environment step; expected recency, how recent sampled transitions are; and the entropy of the replay sampling distribution. Our main contribution is clarifying when non-uniform replay is beneficial and providing practical guidance for replay design in modern off-policy RL. Namely, we find that non-uniform replay is most beneficial when replay volume is low, and that high-entropy sampling is important even at comparable expected recency. Motivated by these findings, we adopt a simple Truncated Geometric replay that biases sampling toward recent experience while preserving high entropy and incurring negligible computational overhead. Across large-scale parallel simulation, single-task, and multi-task settings, including three modern algorithms evaluated on five RL benchmark suites, this replay sampling strategy improves sample efficiency in low-volume regimes while remaining competitive when replay volume is high.

Michal Korniak, Mikołaj Czarnecki, Yarden As +3

1ETH Zurich · University of Warsaw · 3UC Berkeley +1

Deterministic Replay Replay Rejection Sampling

May 7, 2026cs.LG

Revisiting Adam for Streaming Reinforcement Learning

Learning from a sequence of interactions, as soon as observations are perceived and acted upon, without explicitly storing them, holds the promise of simpler, more efficient and adaptive algorithms. For over a decade, however, deep reinforcement learning walked the contrary path, augmenting agents with replay buffers or parallel sampling routines, in an effort to tame learning instability. Recently, this topic has been revisited by Elsayed et al. (2024), focusing on update computation through eligibility traces and modifications to the optimisation routine, resulting in the StreamQ algorithm. In this work we take a step back, investigating the efficacy of established updates, such as those implemented by DQN and C51 within this online setting. Not only do we find that they perform well, but through analysing how the optimisation algorithm generally, and Adam in particular, interacts with these updates, we contend that two properties are essential for robust performance: i) the derivative of the objective is to be bounded and ii) weight updates are variance-adjusted. Rigorous and exhaustive experimentation demonstrates that C51, which exhibits both characteristics, is competitive with StreamQ across a subset of 55 Atari games. Using these insights, we derive a variance-adjusted algorithm based on eligibility traces, termed Adaptive Q

(λ)

, which approaches double the human baseline on the same subset, surpassing existing methods by all performance metrics.

Florin Gogianu, Adrian Catalin Lutu, Razvan Pascanu

Bitdefender, Bucharest, Romania · Mila, Montreal, Canada

Deep Q-Networks Adam Deterministic Replay Modern Data-Streaming Systems Streaming

May 7, 2026stat.ML

Beyond the Independence Assumption: Finite-Sample Guarantees for Deep Q-Learning under $τ$ -Mixing

Finite-sample analyses of deep Q-learning typically treat replayed data as independent, even though it is sampled from temporally dependent state-action trajectories. We study the Deep Q-networks (DQN) algorithm under explicit dependence by modelling the minibatches used for updating the network as

τ

-mixing. We show that this assumption holds under certain dependence conditions on the underlying trajectories and the mechanism used to sample minibatches. Building on this observation, we extend statistical analyses of DQN with fully connected ReLU architectures to dependent data. We formulate each update as a nonparametric regression problem with

τ

-mixing observations and derive finite-sample risk bounds under this dependence structure. Our results show that temporal dependence leads to a degradation in the statistical rate by inducing an additional dimensionality penalty in the rate exponent, reflecting the reduced effective sample size of

τ

-mixing data. Moreover, we derive the sample complexity of DQN under

tau

-mixing from these risk bounds. Finally, we empirically demonstrate on standard Gymnasium environments that the independence assumption is systematically violated and that replay sampling yields approximately exponentially decaying correlations, supporting our theoretical framework.

Leon Halgryn, Sophie Langer, Janusz M. Meylahn +1

University of Twente · Department of Applied Mathematics University of Twente · Ruhr-Universität Bochum +1

Deep Q-Networks Deterministic Replay Conditional Independence Temporal Dependence

Apr 30, 2026cs.CL

Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents

Large language model (LLM)-based coding agents increasingly rely on external memory to reuse prior debugging experience, repair traces, and repository-local operational knowledge. However, retrieved memory is useful only when the current failure is genuinely compatible with a previous one; superficial similarity in stack traces, terminal errors, paths, or configuration symptoms can lead to unsafe memory injection. This paper reframes issue-memory use as a selective, risk-sensitive control problem rather than a pure top-k retrieval problem. We introduce RSCB-MC, a risk-sensitive contextual bandit memory controller that decides whether an agent should use no memory, inject the top resolution, summarize multiple candidates, perform high-precision or high-recall retrieval, abstain, or ask for feedback. The system stores reusable issue knowledge through a pattern-variant-episode schema and converts retrieval evidence into a fixed 16-feature contextual state capturing relevance, uncertainty, structural compatibility, feedback history, false-positive risk, latency, and token cost. Its reward design penalizes false-positive memory injection more strongly than missed reuse, making non-injection and abstention first-class safety actions. In deterministic smoke-scale artifacts, RSCB-MC obtains the strongest non-oracle offline replay success rate, 62.5%, while maintaining a 0.0% false-positive rate. In a bounded 200-case hot-path validation, it reaches 60.5% proxy success with 0.0% false positives and a 331.466 microseconds p95 decision latency. The results show that, for coding-agent memory, the key question is not only which memory is most similar, but whether any retrieved memory is safe enough to influence the debugging trajectory.

Mehmet Iscan

PythaLab, Yildiz Technical University, Istanbul, Türkiye.

Coding Agents Large Language Model Agents Conversational Memory Deterministic Replay

Apr 28, 2026cs.LG

TSN-Affinity: Similarity-Driven Parameter Reuse for Continual Offline Reinforcement Learning

Continual offline reinforcement learning (CORL) aims to learn a sequence of tasks from datasets collected over time while preserving performance on previously learned tasks. This setting corresponds to domains where new tasks arise over time, but adapting the model in live environment interactions is expensive, risky, or impossible. However, CORL inherits the dual difficulty of offline reinforcement learning and adapting while preventing catastrophic forgetting. Replay-based continual learning approaches remain a strong baseline but incur memory overhead and suffer from a distribution mismatch between replayed samples and newly learned policies. At the same time, architectural continual learning methods have shown strong potential in supervised learning but remain underexplored in CORL. In this work, we propose TSN-Affinity, a novel CORL method based on TinySubNetworks and Decision Transformer. The method enables task-specific parameterization and controlled knowledge sharing through a RL-aware reuse strategy that routes tasks according to action compatibility and latent similarity. We evaluate the approach on benchmarks based on Atari games and simulations of manipulation tasks with the Franka Emika Panda robotic arm, covering both discrete and continuous control. Results show strong retention from sparse SubNetworks, with routing further improving multi-task performance. Our findings suggest that similarity-guided architectural reuse is a strong and viable alternative to replay-based strategies in a CORL setting. Our code is available at: https://github.com/anonymized-for-submission123/tsn-affinity.

Dominik Żurek, Kamil Faber, Marcin Pietron +2

AGH University of Krakow, Poland · American University, Washington, DC, USA

Replay-Based Continual Learning Model-Based Reinforcement Learning Offline Reinforcement Learning Deterministic Replay Reuse

Apr 24, 2026cs.CR

Sovereign Agentic Loops: Decoupling AI Reasoning from Execution in Real-World Systems

Large language model (LLM) agents increasingly issue API calls that mutate real systems, yet many current architectures pass stochastic model outputs directly to execution layers. We argue that this coupling creates a safety risk because model correctness, context awareness, and alignment cannot be assumed at execution time. We introduce Sovereign Agentic Loops (SAL), a control-plane architecture in which models emit structured intents with justifications, and the control plane validates those intents against true system state and policy before execution. SAL combines an obfuscation membrane, which limits model access to identity-sensitive state, with a cryptographically linked Evidence Chain for auditability and replay. We formalize SAL and show that, under the stated assumptions, it provides policy-bounded execution, identity isolation, and deterministic replay. In an OpenKedge prototype for cloud infrastructure, SAL blocks 93% of unsafe intents at the policy layer, rejects the remaining 7% via consistency checks, prevents unsafe executions in our benchmark, and adds 12.4 ms median latency.

Jun He, Deying Yu

Agent Loop Deterministic Replay

Apr 23, 2026quant-ph

Replay-buffer engineering for noise-aware quantum circuit optimization

Deep reinforcement learning for quantum circuit optimization faces three bottlenecks: replay buffers that overlook temporal difference (TD) target reliability, curriculum-based architecture search requiring a full quantum-classical evaluation after every edit, and the discard of noiseless trajectories when retraining under hardware noise. We address these limitations by treating replay as a central algorithmic lever. We introduce ReaPER+, an annealed replay rule that transitions from TD-error prioritization to reliability-aware sampling as value estimates mature. ReaPER+ achieves up to 4x higher sample efficiency than fixed PER, ReaPER, and uniform replay, while matching prior on-policy solution quality with up to 32x fewer interactions At 12 qubits, fixed ReaPER reaches the lowest energy error in the fewest steps, while PER and uniform replay find more compact circuits at higher error. On tasks scaling to 20 qubits, ReaPER+ retains its advantage, demonstrating that reliability-aware annealing extends beyond small-system benchmarks. LunarLander-v3 confirms that the ReaPER+ is domain-agnostic, it improves success rates by up to 26.8% over PER and 21.8% over fixed ReaPER, with a 3% AUC gain over both. We further introduce OptCRLQAS, which amortizes quantum-classical evaluations across multiple architectural edits, reducing training wall-clock time by up to 67.5% on 12-qubit without degrading solution quality. Finally, lightweight replay-buffer transfer warm-starts noisy optimization from noiseless trajectories, without weight transfer or

ε

-greedy pretraining, reducing steps to chemical accuracy by 85-90% and final energy error by up to 90% relative to from-scratch learning. Transfer gains increase with system size. Together, these results establish experience storage, sampling, and transfer as decisive levers for sample efficient, noise-aware quantum circuit optimization.

Akash Kundu, Sebastian Feld

Delft University of Technology, Delft, The Netherlands · Quantum Computing Division, QuTech, The Netherlands

Quantum Circuits Hybrid Quantum-Classical Pipeline Deterministic Replay Engineering

Apr 22, 2026cs.AI

Stateless Decision Memory for Enterprise AI Agents

Enterprise deployment of long-horizon decision agents in regulated domains (underwriting, claims adjudication, tax examination) is dominated by retrieval-augmented pipelines despite a decade of increasingly sophisticated stateful memory architectures. We argue this reflects a hidden requirement: regulated deployment is load-bearing on four systems properties (deterministic replay, auditable rationale, multi-tenant isolation, statelessness for horizontal scale), and stateful architectures violate them by construction. We propose Deterministic Projection Memory (DPM): an append-only event log plus one task-conditioned projection at decision time. On ten regulated decisioning cases at three memory budgets, DPM matches summarization-based memory at generous budgets and substantially outperforms it when the budget binds: at a 20x compression ratio, DPM improves factual precision by +0.52 (Cohen's h=1.17, p=0.0014) and reasoning coherence by +0.53 (h=1.13, p=0.0034), paired permutation, n=10. DPM is additionally 7-15x faster at binding budgets, making one LLM call at decision time instead of N. A determinism study of 10 replays per case at temperature zero shows both architectures inherit residual API-level nondeterminism, but the asymmetry is structural: DPM exposes one nondeterministic call; summarization exposes N compounding calls. The audit surface follows the same one-versus-N pattern: DPM logs two LLM calls per decision while summarization logs 83-97 on LongHorizon-Bench. We conclude with TAMS, a practitioner heuristic for architecture selection, and a failure analysis of stateful memory under enterprise operating conditions. The contribution is the argument that statelessness is the load-bearing property explaining enterprise's preference for weaker but replayable retrieval pipelines, and that DPM demonstrates this property is attainable without the decisioning penalty retrieval pays.

Vasundra Srinivasan

Data Engineering for Multimodal AI (O’Reilly) · Stanford School of Engineering

Long-Term Agent Memory Persistent State Decisions Deterministic Replay Artificial Intelligence Agents Penalties

Feb 2, 2026cs.LG

VLM-Guided Experience Replay

Recent advances in Large Language Models (LLMs) and Vision-Language Models (VLMs) have enabled powerful semantic and multimodal reasoning capabilities, creating new opportunities to enhance sample efficiency, high-level planning, and interpretability in reinforcement learning (RL). While prior work has integrated LLMs and VLMs into various components of RL, the replay buffer, a core component for storing and reusing experiences, remains unexplored. We propose addressing this gap by leveraging VLMs to guide the prioritization of experiences in the replay buffer. Our key idea is to use a frozen, pre-trained VLM as an automated evaluator to identify and prioritize promising sub-trajectories from the agent's experiences. Across scenarios, including game-playing and robotics, spanning both discrete and continuous domains, agents trained with our proposed prioritization method achieve 15-57% higher average success rates and improve sample efficiency by 35-55% compared to previous approaches. Project page: https://esharony.me/projects/vlm-rb/

Elad Sharony, Tom Jurgenson, Orr Krupnik +2

Technion · ForSight Robotics · Nvidia Research

Large Language Model Reinforcement Learning Prioritized Experience Replay Deterministic Replay

Jan 17, 2026cs.AI

Replayable Financial Agents: A Determinism-Faithfulness Assurance Harness for Tool-Using LLM Agents

Tool-using agents can repeat a final decision while changing their recorded execution. We introduce the Determinism-Faithfulness Assurance Harness (DFAH), a framework that distinguishes decision repeatability, trajectory agreement, and evidence-conditioned faithfulness. Task correctness requires separately qualified labels and evaluation; evidence-conditioned faithfulness was not evaluated in the historical v2 agentic experiments. The original v2 study reported 4,705 agentic runs in three synthetic financial tasks and a decision-determinism/task-label-match correlation of r = -0.11 across 21 model-benchmark configuration summaries. This statistic is reproducible from the historical configuration table, but includes a subsequently excluded portfolio fixture. It is retained as a historical description, not evidence of statistical independence, predictive uselessness, or an architectural determinism-accuracy tradeoff. Recorded decision concentration and tool-path variation do not identify hidden model strategy. This correction qualifies the historical evidence and removes the deployment recommendations derived from those unsupported interpretations. A separate corrected study, DFAH-Bench (arXiv:2607.20491), provides qualified evidence of decision/path disagreement. The contribution retained here is a measurement framework: repeatability, observable execution, evidence alignment, and correctness require distinct evidence, with explicit capture and study boundaries.

Raffi Khatchadourian

IBM Financial Services Market, New York, NY, USA

Reproducibility Agent Loop Deterministic Replay Language-Model Agents Large Language Model Agents Synthetic Task

Sep 24, 2025cs.LG

Frictional Q-Learning

Off-policy reinforcement learning suffers from extrapolation errors when a learned policy selects actions that are weakly supported in the replay buffer. In this study, we address this issue by drawing an analogy to static friction. From this perspective, the replay buffer is represented as a smooth, low-dimensional action manifold, where the support directions correspond to the tangential component, while the normal component captures the dominant first-order extrapolation error. This decomposition reveals an intrinsic anisotropy in value sensitivity that naturally induces a stability condition analogous to a friction threshold. To mitigate deviations toward unsupported actions, we propose Frictional Q-Learning, an off-policy algorithm that encodes supported actions as tangent directions using a contrastive variational autoencoder. We further show that an orthonormal basis of the orthogonal complement corresponds to normal components under mild local isometry assumptions. Extensive empirical results on standard continuous-control benchmarks consistently demonstrate robust and stable performance compared with competitive baselines.

Hyunwoo Kim, Hyo Kyung Lee

Data Analysis and Healthcare System Lab, Korea University, Seoul, Korea.

Q-Learning Friction Deterministic Replay Extrapolation

Feb 12, 2024stat.ML

Replicability is Asymptotically Free in Multi-armed Bandits

We consider a replicable stochastic multi-armed bandit algorithm that ensures, with high probability, that the algorithm's sequence of actions is not affected by the randomness inherent in the dataset. Replicability allows third parties to reproduce published findings and assists the original researcher in applying standard statistical tests. We observe that existing algorithms require

O(K^2/ρ^2)

times more regret than nonreplicable algorithms, where

K

is the number of arms and

ρ

is the level of nonreplication. However, we demonstrate that this additional cost is unnecessary when the time horizon

T

is sufficiently large for a given

K, ρ

, provided that the magnitude of the confidence bounds is chosen carefully. Therefore, for a large

T

, our algorithm only requires

K^2/ρ^2

times smaller amount of exploration than existing algorithms. To ensure the replicability of the proposed algorithms, we incorporate randomness into their decision-making processes. We propose a principled approach to limiting the probability of nonreplication. This approach elucidates the steps that existing research has implicitly followed. Furthermore, we derive the first lower bound for the two-armed replicable bandit problem, which implies the optimality of the proposed algorithms up to a

\log\log T

factor for the two-armed case.

Junpei Komiyama, Shinji Ito, Yuichi Yoshida +1

New York University · Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) · RIKEN AIP +2

Stochastic Multi-Armed Bandits Reproducibility Deterministic Replay Confidence Intervals Stochastic Exploration

Date pendingcs.RO

Retriever: Composing the Perception-Reasoning-Action Loop for Long-Horizon Manipulation

Building long-horizon robot agents requires composing closed-loop pipelines -- perception, belief update, planning, and control -- whose components run at different clocks and with variable latency. Today, these systems are often assembled with ad-hoc concurrency and pub/sub conventions that make timing and input-consumption semantics implicit, yielding schedule-dependent behavior that is hard to reproduce, debug, and reuse. Current solutions typically solve parts of this problem at either the algorithmic or the systems layer, but not both. In this work, we propose Retriever, which spans the entire stack: an asynchronous decision model, a programming model, a runtime, and an example closed-loop agent pipeline. Retriever represents an agent as a graph of stateful causal stream functions executed on explicit run clocks. We formalize this view via an asynchronous environment-agent loop over continuous-time streams and show that finite-memory causal policies can be represented by compositions of these operators. Retriever compiles these graphs into a runtime that supports multiple backends, enabling systematic debugging across running environments and deterministic replay from logged asynchronous data. We evaluate Retriever through a real-robot case study together with controlled studies of runtime overhead and deterministic replay behavior.

Linfeng Zhao, Haojie Huang, Jiayuan Mao +3

Stanford University · Northeastern University · MIT +1

Agent Loop Deterministic Replay Robot State Robot Systems Retrievers