Skip to content
AI NEWS.ccAI正在发生 未来正在更新
Bookmarks
THE JOURNAL
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

arXiv cs.AI

Decision Context: Persistent Behavioral Learning for AI Agents

arXiv:2611.08644v1 Announce Type: new Abstract: LLM agents do not learn from corrections across sessions. A user fixes the same mistake at the start of every conversation. Existing agent memory systems primarily handle facts, such as preferences, entities, and conversational history, rather than procedural knowledge. We present Forge Decision Context (forge_dc), a system that captures behavioral signals from human corrections and self-assessed tool-call failures, compiles them into imperative RULE directives, and injects them at the decision points where they apply. We evaluate forge_dc alongside four predecessor approaches on DC-Bench, a new benchmark of 71 scenarios across nine behavioral dimensions, built from real executed agent traces in a deterministic mocked-tool environment. forge_dc reaches 95% compliance, up from 55% with naive retrieval, while injecting approximately 10 times fewer tokens per query. Across the five-stage progression, the strongest systems represent behavioral memory as explicit directives and deliver it at the decision points where it applies. The results identify representation, activation, and placement as central design choices for persistent behavioral learning.

arXiv cs.LG

Evolutionary Adaptive Metric Learning: A Multi-Objective Framework for Learning Local Distance Functions in Nearest Neighbor Classification

arXiv:2611.08549v1 Announce Type: new Abstract: Nearest neighbour classifiers inherit the geometry of their distance function. Fixed distances and globally learned metrics apply the same geometry throughout the input space, although relevant feature directions may vary across regions. We propose EAML (Evolutionary Adaptive Metric Learning), a framework that learns instance-dependent positive-definite metrics through a low-dimensional parameterization. Each local metric combines an input-dependent diagonal scaling, generated by softmax interpolation between anchor points, with a global low-rank correlation component. A real-coded elitist genetic algorithm optimizes the parameters using a scalar objective combining leave-one-out prediction error, robustness, metric smoothness, and regularization. We establish positive definiteness, metric equivalence, existence of an optimal solution, and Lipschitz continuity of the metric field. We further investigate prediction stability, generalization bounds, and evolutionary convergence under explicit assumptions. Experiments on synthetic data with region-dependent feature relevance show that EAML can recover local metric variations and outperform global metrics. Evaluation on thirteen public datasets using nested 5-fold cross-validation reveals a more nuanced picture: EAML improves performance on selected datasets but does not significantly outperform global metric baselines overall. Additional experiments show that global diagonal metrics can be preferable under nuisance features and test-time noise. These findings identify the conditions under which evolutionary local metric learning is beneficial and highlight its practical limitations.

arXiv cs.LG

Interpretable Multifidelity Learning with Kolmogorov-Arnold Networks for Spatiotemporal Nuclear Reactor Flow Modeling

arXiv:2611.08512v1 Announce Type: new Abstract: High-fidelity computational fluid dynamics (CFD) simulations of nuclear reactor systems are computationally expensive, motivating machine learning methods that reduce cost while preserving accuracy. Multifidelity surrogate modeling addresses this by combining inexpensive low-fidelity data with limited high-fidelity data. This work investigates whether the high-fidelity data required to predict the spatiotemporal mass flow rate across all 193 fuel assemblies of a pressurized water reactor (PWR) core can be reduced, comparing black-box multifidelity feedforward neural networks (MFNNs) with interpretable multifidelity Kolmogorov-Arnold Networks (MFKANs). Cross-fidelity correlation across the core is weak, making this a challenging multifidelity problem, and neither architecture predicts the transient particularly well in absolute terms. MFKAN nonetheless performs comparably to MFNN for one-time-step-forward prediction while also providing closed-form symbolic equations unavailable with MFNN. To address the scarcity of CFD transients, with only one available per fidelity level, TabDDPM-generated synthetic data was used to augment cross-fidelity forecasting. KAN and FNN achieve comparable self-evaluation accuracy (Pearson correlation of roughly 0.73--0.81), but both perform poorly at true cross-fidelity forecasting (correlations of 0.03--0.06). Assembly-level analysis shows that each assembly's forecast correlation tracks its real cross-fidelity correlation for both architectures ($r=0.840$--$0.896$ for KAN, $r=0.804$--$0.903$ for FNN), confirming that forecasting accuracy is fundamentally limited by how correlated the fidelity levels already are. Overall, KAN-based multifidelity modeling achieves accuracy consistently comparable to MFNN across all settings, while additionally providing symbolic equations that match or exceed the spline representation's accuracy.

arXiv cs.DC

ISER-MoE: Dynamic In-Switch Replication of Hot Experts for Load Balancing in Expert-Parallel

arXiv:2611.09008v1 Announce Type: new Abstract: Expert-parallel mixture-of-experts (MoE) layers run at the pace of the most loaded GPU, and a few hot experts carry most of the excess load. Replicating them at every MoE layer balances the load only if their weights arrive within the layer, and broadcasting from the owner is the bottleneck: point-to-point copies overload it, and baseline broadcasts pay latency or bandwidth overheads. We present ISER-MoE (In-Switch Expert Replication for MoE), built in TensorRT-LLM, which multicasts hot experts through the NVLink switch to fixed GPU groups, so each byte leaves its owner once. A GPU scheduler assigns hot experts to groups per layer, and a multi-port broadcast spreads the injection over the group's ports, both overlapped with the shared expert. In trace replay, the scheduler removed 98-99% of the excess load; in prefill serving of DeepSeek-V4-Pro, throughput rose 17.4% on a math benchmark and 7.2% on agentic prompts.

arXiv cs.AI

Pdb or Not to Be: Comparing Debugging Interfaces With Pass$_{X}$@$k$

arXiv:2611.08666v1 Announce Type: new Abstract: Interactive debuggers are indispensable to human software development, yet coding agents overwhelmingly rely on edit-execute-observe loops, using common bash tools, test scripts, and print statements to debug via repeated shell commands. Stepping through a debugger provides live access to runtime state, which may enable different problem-solving capabilities. However, prior work typically introduces the debugger as an additional tool or through a new subagent, making its individual contribution difficult to isolate from the associated scaffold changes. To address this, we compare a debugger-only agent against a bash-only agent on all 500 SWE-bench Verified instances, holding the rest of the scaffold fixed. We evaluate both architectures on two models, GPT-OSS-120B and Nemotron-3-Ultra-550B, with three independent runs per architecture $\times$ model configuration. Crucially, to distinguish interface specialization from expanded coverage under repeated sampling, we introduce pass$_X$@$k$, a cross-agent pass@$k$ that provides an unbiased estimator of the probability that at least one of $k$ agents is successful. Although the bash and debugger agents solve partially disjoint subsets of the benchmark in any single pair of runs, our pass$_X$@$2$ estimator reveals that this is largely due to run-to-run variance rather than tool complementarity. Across both models, we find that two bash rollouts outperform the combination of one bash and one debugger rollout. Our findings highlight the importance of accounting for stochastic variance in evaluating agent and tool diversity, and we hope that pass$_X$@$k$ will serve as a useful framework for the community.

arXiv cs.CL

Structuring Survey Histories for LLM Digital Twins: A Controlled Study of Background, Decision Procedure, and Evaluation Profiles

arXiv:2611.08730v1 Announce Type: new Abstract: LLM-based digital twins aim to predict how individuals would respond in new settings from representations of their prior responses. Existing evidence that compression of survey histories need not reduce predictive accuracy suggests that representation structure, rather than information volume alone, may be an important design choice. We study BDE, a fixed representation that organizes each respondent's survey history into Background, Decision procedure, and Evaluation. We evaluate it on Twin-2K-500 (Toubia et al., 2025) by predicting each respondent's held-out answers on 17 tasks. In a paired evaluation with 100 respondents, a fully inferred BDE profile improves overall accuracy over the raw survey transcript by 1.27 percentage points (pp) on gpt-5.4-nano (p < 0.01) and 1.19 pp on gpt-5.4-mini (p < 0.001). Three analyses then examine the schema, extraction depth, and component composition. Against a length-matched free-form profile written from the same transcript by the same extractor, BDE improves pricing accuracy on both simulators (2.15 pp on nano, p < 0.01; 1.38 pp on mini, p < 0.05), while its overall advantage is significant only on nano (1.05 pp, p < 0.05). Allowing deeper inference during extraction yields no statistically significant gain on either simulator. By contrast, adding Evaluation to a Background--Decision procedure profile improves overall accuracy by 1.44 pp on nano (p < 0.001) and 1.22 pp on mini (p < 0.001). These results indicate that the advantage over raw transcripts is not explained by compression or length alone, and attribute it primarily to profile composition rather than deeper extraction.

MarkTechPost

Zyphra ZYFA: FP8 FlashAttention for AMD MI355X, Up to 1.76x Faster

Zyphra released ZYFA, open-source FP8 and BF16 prefill attention kernels for AMD Instinct MI355X. On real model activations, its FP8 kernels run 1.16-1.76x faster than AMD’s AITER, with 62-78% lower error at head dimension 128. The post Zyphra ZYFA: FP8 FlashAttention for AMD MI355X, Up to 1.76x Faster appeared first on MarkTechPost.

arXiv cs.AI

Closed-Loop Agent Evaluation: Layered Measurement and Self-Calibration

arXiv:2611.07051v1 Announce Type: new Abstract: Once an agent is live, teams want to know whether it is doing its job, not just whether it throws errors. Yet most evaluation is one-off: a benchmark run, a look at traces, a single score. Its conclusions neither check themselves nor reach the agent. We propose closed-loop evaluation, an evaluation method for deployed agents. It models evaluation as performance estimation under biased raters and censored outcomes, and it requires every level's conclusion to trace back to atomic data and to be computed by versioned rules. The method covers four levels of analysis (span, trace, conversation and engagement), measures along a spine of Execution, Substance, Trajectory and Results, treats Conduct as a constraint throughout, and uses Provenance and Adjudication components to certify the evidence and the evaluators. Evaluator agents generated by an Agent Factory carry these components, each with a domain knowledge pack and an adapter harness. Two loops close the evaluation: an inner loop audits the evidence, rules and metrics themselves, and an outer loop uses per-component regression checks to drive agent improvement. We also give a pipeline that runs this process automatically. Using the business-analysis agent SharpInsight as a case and porting the method to public conversation data, we show how problems differ across levels and how evidence-and-rule judgment is more reliable than holistic scoring (ICC 0.86 vs. 0.81; kappa 0.81 vs. 0.63).

arXiv cs.CL

Information-Matched Auditing of Conversational Emotion-Shift Forecasting: A Leakage-Safe Evaluation Protocol

arXiv:2611.05895v1 Announce Type: new Abstract: Comparisons in conversational emotion-shift forecasting are easily distorted when a learned model receives rich features but its transition baseline is given a noisy or unavailable current-emotion label, when hyperparameter budgets differ, or when target-turn information enters the representation. We present an information-matched, leakage-safe auditing protocol that closes these gaps with an explicit commitment point, executable causality tests, equal four-configuration validation searches, and seed-dialogue-aware inference. Applying it to six causal model families on four text corpora and two feature-source sensitivity configurations shows that a deployable train-only current-emotion classifier feeding the transition baseline yields a different verdict than an oracle diagnostic where both sides receive gold current emotion. Under the deployable condition, learned models improve over the predicted-label baseline on all four primary corpora; DailyDialog gains are large and survive a 72-comparison Holm correction (best ROC-AUC gain = +0.167, adjusted p=.0144). Under matched gold information, the baseline remains competitive on IEMOCAP and stronger on MELD, while every model retains a large DailyDialog advantage. We additionally report precision-recall AUC, calibration, a strictly utterance-local feature control, same-person self-shift forecasting, and a future-access dose-response test. An eight-item leakage checklist accompanies the protocol. Code, raw seed predictions, and reporting artifacts are available from the corresponding author upon request.

arXiv cs.CL

ProcessPrefBench: Evaluating Cross-Task Transfer of Individual Process Preferences from Naturalistic Human-AI Interaction

arXiv:2611.06065v1 Announce Type: new Abstract: In natural human-AI dialogue, users' corrections, requests, and self-descriptions carry reusable signals about how they want work to be done. These signals reveal individual process preferences, user-specific standards for evidence, depth, analytical focus, procedure, and delivery. We ask whether a preference revealed in one dialogue can shape an assistant's later answer when the task does not restate it. ProcessPrefBench uses three months of natural dialogues from two knowledge workers to build 66 source-target cases separately judged for source evidence and target applicability. Sources are preselected, so the benchmark tests application rather than retrieval. We test three models with target-only, full-history, and distilled-preference prompts. Target-specific 0-2 rubrics score preference realization, and a task gate flags severe failures. Compared with target-only prompting, full-history and imperative distilled-preference prompting raise preference realization by 0.204 and 0.244 points on the 0-2 scale, respectively, with no reliable advantage from distillation. Fabricated-evidence failures rise from 5.8% under target-only prompting to 10.1% under full-history and 10.4% under imperative prompting. ProcessPrefBench highlights a tension in learning from dialogue, where drawing on prior interactions can improve preference realization yet also invite fabricated evidence. We hope it helps build assistants that honor a user's process preferences without sacrificing factual reliability.

arXiv cs.CV

PSGAgent: Enhancing MLLMs' Dynamic Visual Perception for Panoptic Scene Graph Generation via Agentic Reasoning

arXiv:2611.07408v1 Announce Type: new Abstract: Panoptic Scene Graph generation (PSG) requires holistic yet fine-grained reasoning over complex relations among objects. However, current Multimodal Large Language Models (MLLMs) applied to PSG suffer from a perception bottleneck: they function as static observers limited by fixed-resolution views. As a result, models often overlook critical relations involving small or peripheral objects and struggle to differentiate ambiguous predicates in complex scenes. To address these limitations, we propose PSGAgent, an agentic framework that enhances visual perception for PSG through a think-act-observe loop. PSGAgent allows MLLMs to dynamically refine their visual perception using two tools: a Region Zoom-In Tool for local exploration of small or boundary relations and a Visual Highlight Tool for relation-focused segmentation that filters out irrelevant contextual noise. To optimize decision-making, we construct a two-stage training pipeline. First, we create a high-quality, error-aware Multimodal Chain-of-Thought (M-CoT) dataset that provides robust supervision for cold-start learning. Then, we design a novel hierarchical reward mechanism integrated with Group Sequence Policy Optimization (GSPO) to enable strategic and adaptive tool usage. Extensive experiments on the PSG benchmark show that PSGAgent achieves SOTA performance, demonstrating strong capability in both holistic and fine-grained scene understanding.

arXiv cs.DC

A Controlled Experimental Evaluation of Configuration and Cache Architecture in LLM Serving Engines

arXiv:2611.05991v1 Announce Type: new Abstract: Multi-agent LLM applications repeat most of their prompt tokens across requests, and serving engines reuse the cached state of shared prefixes. vLLM and SGLang reuse it at different granularities, and benchmarks of their default deployments report large performance gaps, usually credited to the cache. Such comparisons rarely control for other configuration differences. One of them, whether a long prompt is prefilled in chunks, is set by each engine's defaults from the hardware and the release, and it confounds the comparison. We built a reproducible testbed on Kubernetes and a seeded multi-agent workload generator, and compared the two engines under vendor defaults and with chunked prefill matched. SGLang's configuration is the same in both arms, which makes it an internal control across independently provisioned deployments. Under defaults, vLLM's median time to first token is up to 64x SGLang's. Matching one setting removes 78% of the capacity gap where that gap is largest and 30 to 67% of the latency gap. Near saturation the remaining latency gap reflects queue growth. A fluid queue model driven only by a 4.4% throughput shortfall reproduces it, and shows that such a gap scales with the length of the benchmark run. What survives matching is a tail-latency advantage of 2.4 to 6.4x at equal throughput, consistent with cache-aware ordering of waiting requests. We derive six recommendations for comparing serving engines.

arXiv cs.AI

Assurance-Conditioned Continuous Authorization for Autonomous AI Agents

arXiv:2611.05632v1 Announce Type: new Abstract: Autonomous software agents can act through tools, delegate work, and change their models or execution environments while earlier authorizations remain in flight. We present Assurance-Conditioned Continuous Authorization (ACCA), an approach designed to be harness-independent, for maintaining operational authority according to currently justified assurance. It distinguishes evidence validity from contextual applicability, derives assurance claims from eligible evidence, and resolves a multidimensional effective-authority envelope within a principal-declared ceiling and applicable policy. A typed dependency graph supports selective re-evaluation of authorization-relevant changes. An authority epoch changes when effective authority changes; a distinct optional authorization-basis version supports revalidation of action-specific authorization conditions. Protected effect-time enforcement can reject stale actions before a consequential resource change, subject to the available commit mechanism. A compositional Domain Governance Profile expresses domain-specific semantics without granting authority. Controlled scenario tests implemented as a regression suite using a trading profile exercise default-deny rules, evidence-conditioned contraction and recovery, policy divergence, selective non-response, and stale-action rejection at a local gateway before the simulated-broker boundary. The study establishes implementation-level evidence in one domain, not a cross-domain guarantee of harmful-behavior prevention.

arXiv cs.CL

Phonological Knowledge of Chinese Characters in Language Models

arXiv:2611.05803v1 Announce Type: new Abstract: How much phonological knowledge do language models (LMs) possess about Chinese characters, and how is this knowledge represented and used? We introduce PinyinBench, a controlled benchmark covering homophones, phonological features (initials, finals, and tones), and polyphonic characters. We investigate 17 text-only and multimodal LMs through three complementary approaches: behavioral evaluation, layer-wise probing, and activation steering. Behaviorally, models demonstrate substantial knowledge of initials and finals, with a frontier model (DS-v4-flash) scoring 90%+ in homophone judgment, when two characters share no phonological features. However, its accuracy drops sharply to 72% when two characters differ only in lexical tone. Probing reveals that tone is highly decodable from embeddings and improves modestly across layers, whereas segmental information starts less decodable and improves substantially with depth. Activation steering shows that segmental representations can systematically influence generation, while tone exhibits much weaker effects. Furthermore, audio-bearing training improves phonological decodability, including tone, without corresponding improvements in steering effectiveness. Together, these findings reveal distinct representational and functional profiles for segmental and tonal information, with tonal information much harder to access and manipulate, demonstrating the necessity for combined evaluation, probing, and causal intervention for understanding linguistic knowledge in LMs.

arXiv cs.CL

SRRM: A Retrieval-Reconstruction Benchmark for Evaluating Long-Context Memory in Small Language Models

arXiv:2611.05386v1 Announce Type: new Abstract: Small Language Models (SLMs) are increasingly deployed on consumer-grade hardware, yet their ability to preserve information across progressively longer contexts remains insufficiently understood. Existing benchmarks primarily evaluate retrieval-oriented tasks, providing limited insight into whether successful retrieval reflects comprehensive long-context memory preservation. To address this limitation, this paper introduces Simple Retrieval-Reconstruction Memory (SRRM), a lightweight and fully reproducible benchmark that jointly evaluates Targeted Retrieval and Global Information Reconstruction. This study also proposes the Memory Preservation Ratio (MPR), a benchmark-level metric that quantifies the relationship between retrieval and reconstruction performance. SRRM was evaluated using a synthetic dataset on three Qwen2.5 models (0.5B, 1.5B, and 3B parameters) across context loads ranging from 10 to 100 factual statements. Experimental results show that reconstruction performance deteriorated substantially faster than retrieval accuracy as contextual information increased, indicating that retrieval alone is insufficient to characterize long-context memory preservation. Among the evaluated models, Qwen2.5-1.5B achieved the highest retrieval accuracy (86.7%), reconstruction performance (45.7%), and MPR (0.53), outperforming the larger Qwen2.5-3B model (72.2%, 18.5%, and 0.26, respectively). These findings demonstrate that increasing parameter count does not necessarily improve long-context memory preservation. Overall, SRRM provides a reproducible and extensible benchmark for systematically evaluating long-context memory while revealing complementary memory behaviors that remain hidden under retrieval-only evaluation frameworks.

arXiv cs.AI

Towards Efficient Whole Page Ranking with One-Pass Inference

arXiv:2611.05542v1 Announce Type: new Abstract: Whole-page ranking jointly optimizes advertisements and organic content to balance business outcomes and user engagement under complex layout and business constraints. Existing non-autoregressive (NAR) approaches enable efficient page generation but typically score candidate items independently of previously selected items, limiting their ability to capture order-dependent interactions. Autoregressive (AR) methods address this limitation through sequential conditioning, but incur substantial inference overhead that challenges large-scale deployment. Bridging this quality-efficiency gap remains a challenge in industrial whole-page ranking. We propose Multi-Item Scoring (MIS), a prefix-aware NAR-style framework that incorporates sequential context into efficient multi-item scoring. MIS integrates global candidate interactions, localized prefix conditioning, and position-aware scoring within a shared representation architecture. By computing reusable representations once per request and performing lightweight prefix-conditioned scoring during constrained beam search, MIS captures important sequential dependencies without repeated full-model inference. Experiments on KuaiRand-1K and RL4RS demonstrate that MIS approaches the generation quality of an autoregressive CGA baseline while achieving 4.07x and 3.54x faster CPU decoding, respectively. In a randomized evaluation on a large-scale social-video platform, MIS improves the realization metric by 0.104% and watch time by 0.058%, while reducing ranking-service CPU consumption by 9.73%. These results demonstrate that prefix-aware multi-item scoring can narrow the quality gap between NAR and AR generation while retaining the efficiency required for industrial scalability.

arXiv cs.LG

When Latent Dynamics Meet Data Assimilation: A Benchmark of Koopman-Type Representations

arXiv:2611.03916v1 Announce Type: new Abstract: Koopman representations describe nonlinear dynamics through linear evolution in a transformed space, offering an attractive framework for their linear simplicity and generality. Yet these advantages do not necessarily translate into improved performance in deep latent data assimilation (LDA). We systematically compare seven representative methods and structural baselines across five systems spanning approximately periodic to strongly chaotic dynamics and a broad range of state dimensions. We evaluate forecast and assimilation accuracy, physical state recovery, latent-trajectory behavior, and dynamical consistency, and assess Gaussian error assumptions, objective convexity, and uniqueness of analysis solutions. Our results reveal a trade-off between representational flexibility and optimization tractability. Linear evolution of the latent state can stabilize short-term forecasts but restrict expressiveness in strongly chaotic and high-dimensional systems, without consistently improving assimilation accuracy. Parameterized linear operators offer greater flexibility, yet their joint optimization with the encoder--decoder remains nonconvex. Methods with closed-form analysis updates generally perform better and exhibit greater robustness in our experiments, although matrix inversion remains a computational bottleneck. These findings show that the practical benefits of Koopman structure depend on both sufficient observational information and the choice of assimilation procedure.

arXiv cs.CV

Do We Still Need Rotation-Invariant Architectures? A Statistical Robustness Benchmark for Point Cloud Classification

arXiv:2611.03952v1 Announce Type: new Abstract: A substantial line of work builds rotation invariance directly into point cloud architectures. We benchmark 10 architectures (PointNet, PointNet++, DGCNN, RIConv, RIConv++, RISurConv, PointMLP, PointTransformer, and the equivariant VN-PointNet and VN-DGCNN) across 5 domains under pose, noise, and occlusion, and analyze the results with confirmatory hypothesis tests rather than point estimates. The three rotation-invariant (RI) architectures achieve the best pose robustness but rank among the lowest on noise robustness in 4 of 5 domains. We ask whether this cost is intrinsic to invariance or specific to how invariance is obtained. We retrain all 10 architectures unchanged, with a simple invariance loss on the embedding and three training seeds each, and evaluate all models on the independently designed ModelNet40-C. Invariance training transfers narrowly but reliably. It improves exactly the corruption families its perturbations target wherever the baseline has headroom (mean occlusion +10.5, lidar +13.4 percentage points of accuracy). At matched training budget the best standard backbone and the best RI architecture end on par on unseen corruptions (within 2.5 percentage points in every seed), and invariance-trained RIConv++ is the most seed-stable top model. Once rotation augmentation is widened, standard backbones concede only a flat, angle-independent accuracy offset under full-sphere rotation. We release the benchmark along with a package for analyzing point cloud robustness and reproducing our results.

arXiv cs.LG

Does the Trust Boundary Hold over Time? A Long-Horizon Benchmark for Receiver-Directed Federated Learning

arXiv:2611.03724v1 Announce Type: new Abstract: In receiver-directed federated learning, each site draws a trust boundary. For every source and model component it judges, on its own data, whether the transfer helps (admit), harms (reject) or cannot yet be told (abstain). We ask whether this boundary holds over hundreds of rounds. Our benchmark has a federation of thirteen semi-synthetic sites, an exact oracle for every evaluated transfer, paired seeds, an attack protocol, fourteen methods over 600 rounds and Qwen3-1.7B LoRA sites over 500. The answer depends on which boundary is measured. Across 20 paired CPU seeds, a gate judging against a fixed reference better identified transfers useful relative to the site's starting model. Yet it admitted 29% of transfers harmful to the model the site now has. A gate judging against the current model admitted 2% of harmful transfers on that target in the CPU tier, but missed many transfers useful to the starting model. Two short propositions explain these distinct failures. We also evaluate a held out audit that leaves gate verdicts intact. After training, it decides whether each final model is deployed or the site falls back to local training, and reports admissions that have gone stale. With independent audit data its guarantees hold after any training history. At 600 rounds the audit deployed 65 to 72% of models, none with negative benefit on its target, and kept 98% of the gain. No stale report had the wrong sign in our runs. With 20,000 audit observations, a variance-adaptive bound found 99% of stale transfers that cost at least 0.01 in log loss. A common-target Qwen3 comparison on 60 selected transfers shows the same tension over 100 rounds. With 38% adversarial sites, FedFomo and CS-Gate were the only evaluated methods with positive mean gain in every honest group.