arXiv:2606.25996v3 Announce Type: replace Abstract: We introduce Autodata, a general method that enables AI agents to act as data scientists who build high quality training and evaluation data. We show how to train (meta-optimize) such a data scientist agent, so that it learns to create even stronger data. We describe the overall formulation, and a specific practical implementation, Agentic Self-Instruct. We conduct experiments on computer science research tasks, legal reasoning tasks and reasoning with mathematical objects, where we obtain improved results compared to classical synthetic dataset creation methods. Further, meta-optimizing the data scientist agent itself delivers an even larger performance uplift. Agentic data creation provides a way to convert increased inference compute into higher quality model training. Overall, we believe this direction has the potential to change the way we build AI data.
Science Journals
arXiv:2601.20749v2 Announce Type: replace Abstract: How do students develop AI literacy through everyday practice rather than formal instruction? While normative AI literacy frameworks proliferate, empirical understanding of how students actually learn to work with generative AI remains limited. This study analyzes 10,536 ChatGPT messages from 36 undergraduates over one academic year, revealing five use genres -- academic workhorse, emotional companion, metacognitive partner, repair and negotiation, and trust calibration -- that constitute distinct configurations of student-AI learning. Drawing on domestication theory and emerging frameworks for AI literacy, we demonstrate that functional AI competence emerges through ongoing relational negotiation rather than one-time adoption. Students develop sophisticated genre portfolios, strategically matching interaction patterns to learning needs while exercising critical judgment about AI limitations. Notably, repair work during AI breakdowns produces substantial learning about AI capabilities, developing what we term "repair literacy" -- a crucial but underexplored dimension of AI competence. Our findings offer educators empirically grounded insights into how students actually learn to work with generative AI, with implications for AI literacy pedagogy, responsible AI integration, and the design of AI-enabled learning environments that support student agency.
arXiv:2606.26025v3 Announce Type: replace Abstract: Modern Vision-Language-Action (VLA) models often fail to generalize to novel setups, such as altered camera viewpoints or robot morphologies, because they are typically conditioned only on current observations and language instructions. By ignoring the underlying system configuration as a variable, these models implicitly assume a fixed execution context encountered during training, necessitating data-intensive fine-tuning for any new environment. In this work, we introduce In-Context World Modeling (ICWM), a framework that treats system identification as an in-context adaptation problem. ICWM enables robot policies to autonomously infer essential system variables from a short history of self-generated, task-agnostic interactions. Unlike traditional In-Context Learning that uses demonstrations to specify what task to perform, ICWM leverages the context window to understand how the system operates. By processing these interactions before task execution, the model implicitly captures the world dynamics of the current system, enabling adaptation to novel configurations without parameter updates. Extensive experiments in simulation and on real-world robot platforms demonstrate that ICWM significantly outperforms standard VLA baselines on novel camera viewpoints.
arXiv:2602.04853v2 Announce Type: replace Abstract: Large language models often struggle to recognize their knowledge limits in closed-book question answering, leading to confident hallucinations. While decomposed prompting is typically used to improve accuracy, we investigate its impact on reliability. We evaluate three task-equivalent prompting regimes: Direct, Assistive, and Incremental, across different model scales and multi-hop QA benchmarks. We find that although accuracy gains from decomposition diminish in frontier models, disagreements between prompting regimes remain highly indicative of potential errors. Because factual knowledge is typically stable while hallucinations are stochastic, cross-regime agreement provides a precise signal of internal uncertainty. We leverage this signal to implement a training-free abstention policy that requires no retrieval or fine-tuning. Our results show that disagreement-based abstention outperforms standard uncertainty baselines as an error detector, improving both F1 and AUROC across settings. This demonstrates that decomposition-based prompting can serve as a practical diagnostic probe for model reliability in closed-book QA.
arXiv:2602.06205v2 Announce Type: replace Abstract: The Platonic Representation Hypothesis suggests that independently trained neural networks converge to increasingly similar latent spaces. However, current strategies for mapping these representations are inherently pairwise, scaling quadratically with the number of models and failing to yield a consistent global reference. In this paper, we study the alignment of $M \ge 3$ models. We first adapt Generalized Procrustes Analysis (GPA) to construct a shared orthogonal universe that preserves the internal geometry essential for tasks like model stitching. We then show that strict isometric alignment is suboptimal for retrieval, where agreement-maximizing methods like Canonical Correlation Analysis (CCA) typically prevail. To bridge this gap, we finally propose Geometry-Corrected Procrustes Alignment (GCPA), which establishes a robust GPA-based universe followed by a post-hoc correction for directional mismatch. Extensive experiments demonstrate that GCPA consistently improves any-to-any retrieval while retaining a practical shared reference space.
arXiv:2607.04422v1 Announce Type: new Abstract: Recent NVFP4 pretraining methods mainly target transformer linear layers, leaving optimizer states, optimizer arithmetic and attention underexplored in 4-bit pipelines. This critical gap blocks stable full-stack 4-bit pretraining, as the three core modules exhibit unique numerical failure patterns: linear layers hit hard quantization noise limits with dimension-propagated error amplification; AdamW second moments are heavy-tailed non-negative values fragile to low-precision denominators; attention carries error-prone computation paths demanding strict forward-backward quantization consistency. We propose Full-Stack FP4, the first complete NVFP4 pretraining framework resolving all three stability bottlenecks via module-wise precision strategies. For linear projections, LoRA-SVD lightweight decomposition suppresses quantization noise and breaks the direct-quantization error ceiling, shrinking the linear-only loss gap from 1.40% to 0.61%. For optimizers, we design AdamW second-moment transformation for robust NVFP4 storage and fully support native NVFP4 Newton-Schulz iterations for the Root (Muon) optimizer. For attention, a mixed-precision scheme quantizes Q/K/V and backward dS while guarding vulnerable paths in BF16, paired with unified tensor reuse to sustain forward-backward alignment. We further analyze fast error accumulation in naive low-bit matrix multiplication and the extreme sensitivity of PV / dOV^T attention branches. All modules are plug-and-play with cumulative stability and efficiency improvements. Our 3B/64B-token pretraining validates near-BF16 performance with merely 1.47% loss gap, verifying feasible stable end-to-end NVFP4 LLM pretraining.
arXiv:2607.05250v1 Announce Type: new Abstract: Neural speech codecs are increasingly used as intermediate representations in codec-based speech generation systems. TiCodec introduces a factorized representation that separates time-varying speech content from time-invariant information through a Time-Invariant Representation Extraction (TIRE) module, potentially reducing the amount of information that must be modeled at the frame-level. In this work, we investigate the nature of the information captured by TIRE representations and their suitability for low-latency speech processing. Using a series of probing tasks, we analyze the influence of the encoder layer and show that intermediate layers capture complementary speaker- and environment-related information while containing little linguistic content. We further study several segment selection strategies for TIRE training and demonstrate that cross-file sampling improves the robustness of invariant representations. Based on these findings, we propose Dual-TIRE, a multi-level architecture that exploits the complementarity of different encoder layers and improves speech reconstruction quality and speaker similarity. Finally, we evaluate TiCodec in a streaming inference setting using successive 660ms processing blocks. Results show that streaming operation can be achieved without significant degradation in reconstruction performance, highlighting the potential of factorized neural codec representations for future low-latency speech generation systems.
arXiv:2601.16806v4 Announce Type: replace Abstract: Insect neuroethology provides a compelling biological template for efficient autonomous navigation. We draw an analogy between the formal embodied AI visual point-goal navigation task and the ability of insects to discover, learn, and refine visually guided paths around obstacles between a discovered food location and their nest. We develop a novel integrative model of mushroom body and central complex, two insect brain structures, that have been implicated, respectively, in associative learning and path integration. We demonstrate the mushroom body learning triggered by collisions results in adaptive obstacle avoidance and consequently optimised paths to the goal, corroborating the hypothesis of recent behavioural work that an insect can learn continuously as they travel. The embodied insect-inspired model achieves success rates comparable to recent state-of-the-art models at many orders of magnitude less computational cost in the standardised Habitat point-goal navigation benchmark. Testing in a more realistic simulated environment validates its robustness to perturbations.
arXiv:2606.26564v3 Announce Type: replace Abstract: The classical existence result of Nash guarantees that every finite noncooperative game admits an equilibrium in mixed strategies, but it leaves open the question of when pure strategy equilibria exist. This paper develops a structural approach to that question by exploiting properties of the best-response correspondence on finite strategy sets. Building on recent work, we derive new sufficient conditions for the existence of pure strategy Nash equilibria in finite games. We introduce several broad classes of finite games for which pure equilibria are guaranteed, including a class of generalized potential games that generalizes unilaterally competitive games and a class characterized by the existence of an aggregate-payoff maximizer over an ordered set. Our results clarify the role of acyclicity, and aggregation in producing pure equilibria and connect disparate sufficient-condition results in the literature into a unified framework.
arXiv:2606.26997v2 Announce Type: replace Abstract: Large language model (LLM) post-training for reasoning increasingly relies on reinforcement learning with verifiable rewards (RLVR), where models learn from ground-truth feedback on mathematical, logical, and scientific tasks. To enable flexible resource allocation and support heterogeneous training setups, modern RLVR systems adopt disaggregated architectures that decouple rollout generation and policy training across independent GPU pools. However, existing synchronous on-policy GRPO (Group Relative Policy Optimization) RLVR systems finish an entire rollout before starting training, leaving the trainer GPU pool idle while rollout is still ongoing. Asynchronous RL pipelines overlap the two stages, but at the cost of training on stale data. To address these challenges, we propose RolloutPipe, a post-training framework for disaggregated RLVR systems, which turns the fixed-weight rollout into a complete-group pipeline where trainable groups move to the trainer while later groups are still being generated. RolloutPipe achieves this through two techniques including complete-group pipelining (CGP) and frontier-group dispatch (FGD). CGP dispatches each trainable complete group to the trainer FIFO as soon as group materialization finishes, and FGD is an admission policy on the Rollout node that first admits requests for the frontier groups needed to form the next training batch, so that trainer-ready groups arrive earlier and more steadily. The design starts training before the rollout completes while maintaining on-policy correctness. Evaluated on Qwen3-1.7B across four reasoning and science benchmarks and twelve rollout settings, RolloutPipe shortens the rollout-to-train-end time by 30.7%-42.3%, and lowers the trainer waiting ratio by 37%-76% compared to Slime, a state-of-the-art rollout and training system.
arXiv:2606.27112v2 Announce Type: replace Abstract: This paper proposes a corrected heavy-ball Q-learning method for reinforcement learning (RL) and establishes convergence of its deterministic mean dynamics. It also identifies conditions under which the method is theoretically guaranteed to converge faster than standard Q-learning. The same construction is then extended to Q-learning with linear function approximation, where analogous convergence and acceleration statements are derived for the corresponding corrected fixed point. The sampled stochastic versions are treated through conditional-mean recursions and, in the stated linear-function-approximation setting, finite-time bounds. The analysis is based on a switched linear system (SLS) representation of Q-learning algorithms and on the joint spectral radius (JSR) of the associated switching families. This SLS viewpoint is not commonly used in standard analyses of Q-learning, and it provides a complementary framework and new insight into how heavy-ball momentum can accelerate Q-learning.
arXiv:2607.04553v1 Announce Type: new Abstract: We present a bidirectional framework for estimating the energy consumption of text-to-video (T2V) and text-to-video-audio (T2VA) models from architectural first principles and observable generation parameters such as resolution and duration, requiring no access to weights, model size, or implementation details. Forward, it predicts energy from generation parameters and architectural principles; backward, it recovers architectural scaling behavior from observed inference times, with accuracy serving as a criterion for architectural validity. Building on the established compute-bound nature of video diffusion models, we demonstrate that each model's energy profile obeys theoretically derived scaling laws, decomposing into quadratic and linear terms whose coefficients directly reflect the underlying architectural complexity. Validated across six open-source models spanning 8.3B-27B parameters and three GPU configurations, this decomposition achieves below 3% MAPE across all architectures. This approach offers a standardized, empirically and theoretically grounded framework for sustainability benchmarking across T2V models and architectures.
arXiv:2607.03881v1 Announce Type: cross Abstract: Codon harmonization aims to adapt the coding sequences for heterologous expression while preserving the native-like patterns of frequent and rare codons that may influence local translation dynamics and co-translational protein folding. However, widely used harmonization metrics, such as $\%$MinMax, are defined on discrete codon sequences and are, therefore, not readily compatible with gradient-based neural codon design. Here, we introduce Smooth $\%$MinMax, denoted as $\%{\rm MinMax}_{(s)}$, a differentiable relaxation of the conventional hard $\%$MinMax metric, denoted as $\%{\rm MinMax}_{(h)}$. $\%{\rm MinMax}_{(s)}$ replaces the discrete codon-usage values with probability-weighted synonymous-codon usage values and replaces the hard $\%$Max/$\%$Min branch with a sigmoid-gated interpolation. This formulation preserves the signed interpretation of $\%{\rm MinMax}_{(h)}$, while enabling optimization with respect to the synonymous-codon probabilities and learnable parameters. In human-to-Escherichia coli codon harmonization experiments, $\%{\rm MinMax}_{(s)}$ closely approximates $\%{\rm MinMax}_{(h)}$ and supports gradient-based profile matching in synonymous-codon probability space. These results suggest $\%{\rm MinMax}_{(s)}$ as a practical bridge between profile-based codon harmonization and neural synonymous-sequence design.
arXiv:2410.15135v5 Announce Type: replace Abstract: With the surge of online misinformation, Large Language Models (LLMs) and Reasoning Large Language Models (RLMs) serving as Automatic Fact-Checking (AFC) systems have emerged as a prominent paradigm for reliable, explainable verification. However, our empirical study reveals that this paradigm faces a critical risk asymmetry challenge when deployed in the real world under resource-constrained environments. While Hotspot Perception Ability (HPA), the capacity to dynamically allocate reasoning resources based on social impact, is essential to mitigate this risk, existing benchmarks lack the social metadata and evaluation framework to meet this urgent evaluation needs, thereby hindering the advancement of these AFC systems. To bridge this gap, we introduce TrendFact, the first benchmark capable of evaluating HPA and three fact-checking tasks. It consists of 7,643 curated samples sourced from trending platforms and professional datasets, with an evidence library containing 366,634 entries. To enable HPA assessment, we propose two novel metrics: the Explanation Consistency Score (ECS) to evaluate the reliability of verification reasoning, and the Hotspot Claim Perception Index (HCPI) to quantify the overall HPA of AFC systems. Extensive experiments demonstrate that existing AFC systems exhibit limited performance on TrendFact. Furthermore, our proposed FactISR framework effectively enhances HPA and computational efficiency for RLMs-served AFC systems.
arXiv:2606.27205v2 Announce Type: replace Abstract: Large Language Models (LLMs) are powerful tools and have been increasingly adopted for complex software engineering tasks. As the number of parameters increases, results can often be improved, but this also imposes substantial memory requirements. While quantization effectively reduces the memory footprint, its overall impact is often summarized only by benchmark scores, which mask changes in model behavior and non-functional overheads. In this work, we conduct an empirical evaluation of LLM quantization using Automated Program Repair (APR), a complex task in software engineering. We analyze 13 quantization configurations spanning different bit-widths, methods, and target components (weights and KV-cache) across six representative LLMs, evaluated on two APR benchmarks (HumanEval-Java and Defects4J). Our findings reveal that base and quantized models can provide different sets of repaired problems with little overlap, while retaining a comparable number of repaired problems. Although quantization successfully reduces memory footprints by up to 85%, it increases both inference time and energy consumption, which we attribute to suboptimal hardware utilization. Our Pareto trade-off analysis shows that 48% of the configurations evaluated are strictly dominated by alternatives. Rather than identifying a superior quantization method, our findings highlight that the trade-offs between effectiveness, memory footprint, and energy efficiency are sensitive to the underlying model architecture and the complexity of the task.
arXiv:2606.27571v2 Announce Type: replace Abstract: In recent decades, driven by rapid data growth, organisations have faced the need to restructure their storage frameworks to efficiently handle queries requested by employees through available enterprise applications. Investigating this need involves examining the classic approaches of William H. Inmon, widely known as the father of Data Warehousing, and Ralph Kimball. Although both shared the same core concerns, Kimball later suggested an alternative architecture focused primarily on user needs. According to Ariyachandra and Watson, Inmon's "hub-and-spoke" architecture and Kimball's data bus framework featuring conformed dimensions stand out among alternative approaches. A comparison across these architectures highlights four key aspects: Information Quality, System Quality, Individual Impacts, and Organisational Impacts. Although Kimball and Inmon proposed contrasting solutions, they did not view each other as rivals. For instance, in one of the editions of the book "The Data Warehouse Toolkit", published by Kimball in 1996, the back cover features a note by Inmon stating that it is "one of the definitive books of our industry. If you take time to read only one professional book, make it this book.
arXiv:2508.21571v2 Announce Type: replace Abstract: Physics informed neural networks (PINNs) represent a very popular class of neural solvers for partial differential equations. In practice, one often employs stochastic gradient descent type algorithms to train the neural network. Therefore, the convergence guarantee of stochastic gradient descent is of fundamental importance. In this work, we establish the linear convergence of stochastic gradient descent / flow in training over-parameterized two layer PINNs with a general class of activation functions for solving one model second-order elliptic problem, i.e., the Poisson equation, in the sense of high probability. These results extend the existing result [20] in which gradient descent was analyzed. The challenge of the analysis lies in handling the dynamic randomness introduced by stochastic optimization methods. The key of the analysis lies in ensuring the positive definiteness of suitable Gram matrices during the training. The analysis sheds insight into the dynamics of the optimization process, and provides guarantees on physics informed neural networks trained by stochastic algorithms.
arXiv:2510.15686v2 Announce Type: replace Abstract: Learning coordinated behaviors for multi-robot systems from only a few demonstrations is difficult because temporal task dependencies and spatial trajectory generation are tightly coupled, which increases the hypothesis space and often yields unstable generalization in data-scarce regimes. We present DDACE, a structured few-shot learning framework that introduces a structural inductive bias by explicitly decoupling temporal coordination from spatial trajectory synthesis. Demonstrations are first processed via spectral clustering to extract coordination structure and form interaction graphs. A Temporal Graph Network predicts action dependencies and sequences, while Gaussian Process models generate progress-parameterized geometric trajectories that adapt to new start/goal configurations. This factorized design reduces hypothesis coupling and improves data efficiency for few-shot multi-robot coordination. Extensive simulation studies and real-robot experiments show that DDACE produces stable coordinated executions from a small number of demonstrations and improves trajectory consistency compared to end-to-end imitation baselines under limited data. Additional materials are available at https://sites.google.com/view/ddace.
arXiv:2509.06774v3 Announce Type: replace Abstract: Organizations and educational institutions use time-bound assessment tasks to evaluate coding and problem-solving skills. These assessments measure not only the correctness of the solutions, but also their efficiency. Problem setters (educator/interviewer) are responsible for crafting these challenges, carefully balancing difficulty and relevance to create meaningful evaluation experiences. Conversely, problem solvers (student/interviewee) apply critical and logical thinking to arrive at correct solutions. In the era of Large Language Models (LLMs), LLMs assist problem setters in generating diverse and challenging questions, but they can undermine assessment integrity for problem solvers by providing easy access to solutions. We introduce OpenCoderRank, a lightweight, self-hosted platform that emulates real-world timed technical assessments in resource-constrained environments. OpenCoderRank is intentionally model-agnostic: it facilitates the creation, deployment and automatic grading of problems while offering fine-grained control over time limits, input-output pairs and execution constraints. OpenCoderRank is evaluated using two methods: 1. BERTScore, 2. LLM evaluation. Our findings indicate that OpenCoderRank connects problem setters and solvers by supporting time-constrained preparation and self-hosted, customizable assessments in resource-constrained settings.
arXiv:2607.04562v1 Announce Type: new Abstract: Large language models (LLMs) generate fluent outputs that can be wrong. Unlike humans, who often exhibit cues when providing false information, LLMs produce errors that are difficult to detect because autoregressive decoding provides no mechanism for verifying intermediate reasoning before state progression. We introduce Heaviside Continuity of Rolling Coefficients (HCRC), a verification-first execution framework that reformulates inference as predicate-gated state transitions governed by a Heaviside Gate. HCRC combines model confidence with independent verification signals from a parallel worker architecture, allowing execution to advance only when predefined correctness predicates are satisfied. This prevents invalid intermediate states from propagating, reducing epistemic entropy without modifying the underlying model. We evaluate HCRC on software-engineering and reasoning tasks across thirteen proposers from four providers. On capable proposers, the gate reduces the false-completion rate (FCR) from 4--7% to 0% while remaining latency-competitive and, in some settings, faster than the unwrapped model. On weaker proposers, it converts false completions into honest halts instead of corrupting downstream state. Beyond benchmarking, HCRC has operated for months as the production control plane of an agentic coding environment, authorizing file mutations, verification-driven progress reporting, and memory compaction. These results establish HCRC as a general framework for verification-driven LLM execution, showing that reliable reasoning can be achieved through principled execution control rather than model scale alone.
arXiv:2607.04563v1 Announce Type: new Abstract: As language modeling technology matures, there is an increasing research focus on the composition and curation of datasets used to train these models. For instance, practitioners commonly seek to augment high-quality datasets with additional text to enhance the performance of models trained on that data. However, informed decisions about data augmentation require more nuanced assessments about data quality. We build on work measuring the precision and recall of generative models to develop a pair of metrics that quantify (1) fidelity, capturing how closely candidate text resembles reference data, and (2) diversity, capturing how well it covers the modes of the reference dataset. Our metrics are based on optimal transport divergence functionals between discrete text summaries. In experiments on M2D2 text datasets, we show that these metrics are able to disentangle a lack of fidelity from a lack of diversity in deficient candidate text. In further experiments, our metrics detect diversity deficits in synthetic GSM8K-style math datasets, which correlate with degradations in downstream accuracy of language models finetuned on this synthetic data.
arXiv:2509.15206v3 Announce Type: replace Abstract: The high memory demands of generative language models have drawn attention to quantization, which reduces memory usage by mapping model weights to lower-precision integers. However, recent empirical studies show that, while efficient, quantization can increase the likelihood of generating biased outputs and degrade performance on fairness benchmarks. In this work, we draw new links between quantization and model fairness by adding explicit group-fairness constraints to the quantization objective and introduce Fair-GPTQ, the first quantization method explicitly designed to reduce unfairness in large language models. The added constraints guide the learning of the rounding operation toward less-biased text generation for protected groups. Specifically, we focus on stereotype generation involving occupational bias and discriminatory language spanning gender, race, and religion. Fair-GPTQ has minimal impact on performance, preserving at least 90% of baseline accuracy on zero-shot benchmarks, reduces unfairness relative to a half-precision model, and retains the memory and speed benefits of 4-bit quantization.
arXiv:2604.16926v2 Announce Type: replace Abstract: Electroencephalography (EEG) foundation models have shown strong potential for learning generalizable representations from large-scale neural data, yet their clinical deployment is hindered by distribution shifts across clinical settings, devices, and populations. Test-time adaptation (TTA) offers a promising solution by enabling models to adapt to unlabeled target data during inference without access to source data, a valuable property in healthcare settings constrained by privacy regulations and limited labeled data. However, its effectiveness for EEG remains largely underexplored. In this work, we introduce NeuroAdapt-Bench, a systematic benchmark for evaluating test-time adaptation methods on EEG foundation models under realistic distribution shifts. We evaluate representative TTA approaches from other domains across multiple pretrained foundation models, diverse downstream tasks, and heterogeneous datasets spanning in-distribution, out-of-distribution, and extreme modality shifts (e.g., Ear-EEG). Our results show that standard TTA methods yield inconsistent gains and often degrade performance, with gradient-based approaches particularly prone to heavy degradation. In contrast, optimization-free methods demonstrate greater stability and more reliable improvements. These findings highlight the limitations of existing TTA techniques in EEG, provide guidance for future development, and underscore the need for domain-specific adaptation strategies.
arXiv:2606.27780v2 Announce Type: replace Abstract: World models are increasingly used for planning, yet most analyses of rollout error assume vector-valued states and scalar error amplification. Many planning environments, however, are naturally graph-structured: agents, tools, skills, routes, and dependencies interact through evolving relations. In this work, we study how prediction errors accumulate in Graph World Models (GWMs). We formulate fixed-edge and dynamic-edge GWM rollouts under a unified state-action transition framework and derive topology-aware error bounds. For fixed-edge rollouts, we show that long-horizon node error separates into a topology factor, governed by the graph spectral radius, and a model factor, governed by layer spectral norms. For dynamic-edge rollouts, we introduce a joint node-edge error operator that captures feedback between feature prediction and structure prediction, revealing when edge errors amplify future message passing. Motivated by these bounds, we propose Error-Aware GWM, a training objective that combines spectral regularization, rollout consistency, and critical-node weighting. Across synthetic graph topologies and heterogeneous agent-graph testbeds, we find that rollout error and planning regret grow with horizon, that dynamic-edge training is necessary when structure evolves, and that Error-Aware GWM improves long-horizon stability without sacrificing one-step accuracy. Our results characterize when graph world models remain reliable under autoregressive planning and when topology makes them fail.
arXiv:2606.27803v2 Announce Type: replace Abstract: Fuzzy Labeled Private Set Intersection (FLPSI) lets a receiver learn the labels of enrolled records similar to its query, and nothing else. Constructions based on a set-threshold reduction reach practical performance: a query matches a record when the two agree on a threshold number of components, and the private matching is delegated to an inner set-threshold kernel. We study its homomorphic form, which combines leveled-BFV homomorphic encryption (HE), a garbled circuit, and secret sharing to decide the match under encryption and release the record's label. We identify a composition gap in this kernel: efficiency is bought with a per-trial false-accept probability, but one query runs a trial for every record, so the error compounds with the database size into the kernel's realization soundness error (RSE), the rate at which it accepts a query the plaintext matcher would reject. The RSE is a reliability property of the cryptographic matching layer, not the matcher's accuracy, and a sound kernel must contribute zero or negligible RSE of its own. We formalize this as a composable security property, give a closed-form bound on the receiver's advantage, and close the gap with CSTPSI, a kernel that runs independent token rounds and raises the per-trial bound to a matching power. We prove CSTPSI secure in the semi-honest model. The bound sets the round count: two token rounds suffice for million-scale databases and three for billion-scale at the $10^{-6}$ engineering threshold. Our evaluation confirms this: at a million records the baseline kernel's RSE reaches 100% while CSTPSI holds it at 0 in every measured configuration. For large labels at small to moderate scale CSTPSI is more than 20x faster than the baseline, with up to 93% less communication, converging to the baseline only at million-scale. Our implementation, with a one-command reproducibility harness, is publicly available.