# Research Papers — 2026-07 · bundle page 212 of 222

> Full-text markdown of papers #10551–10600 of 11075 published in 2026-07, newest first. AI-readable mirrors of the full 11075-paper 2026-07 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-07/page/212
[← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/211.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-07/page/213.md)

### Scaling Trends for Lie Detector Oversight in Preference Learning

Authors: Oskar J. Hollinsworth, Ann-Kathrin Dombrowski, Sam Adam-Day, Adam Gleave, Chris Cundy
Published: 2026-07-02
Categories: cs.AI
arXiv: https://arxiv.org/abs/2607.01567v1

Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers. In this paper, we scale SOLiD to larger models and evaluate it in more diverse and realistic preference-learning settings. We find favorable scaling: undetected deception drops from 34% for 1B-parameter models to 14% for 405B-parameter models at a detector true positive rate of 99%, and expensive human labelers can be removed entirely from the fine-tuning phase without a statistically significant increase in deception. However, SOLiD is sensitive to distribution shift between detector training and preference-training data, which can drive detector false positive rates to impractical levels.

---

### DiPS: Dialogue Policy Selection for High-Stakes Persuasion Agents

Authors: Tianyi Zhang, Mousumi Das, Abrar Anwar, Jesse Thomason, David Traum
Published: 2026-07-02
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2607.01557v2

Large Language Models (LLMs) often struggle with persuasion in high-stakes scenarios. People's individual personalities and concerns require tailored strategies rather than a one-size-fits-all approach. To address this challenge, we focus on a fire-rescue scenario in which an operator must persuade a resident to evacuate as a high-stakes persuasion domain and propose Dialogue Policy Selection (DiPS), a Q-learning framework to dynamically select persuasion strategies adapted to the evolving conversational context. Specifically, we train a critic, trained to maximize the chance of evacuation success, to select a persuasion policy at each turn based on the resident's recent utterances. We then evaluate DiPS against multiple baselines in both simulated and real human interactions. We find that DiPS achieves higher evacuation success than a zero-shot LLM and generic RAG-augmented approach.

---

### Mind the Gap: Standard 3DGS Evaluation Primarily Measures Near-Trajectory Interpolation

Authors: Gaoxiang Jia, Vikram Appia
Published: 2026-07-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2607.01556v1

Standard MipNeRF360-style 3D Gaussian Splatting (3DGS) evaluation holds out every N-th frame -- but these frames have trained neighbors on both sides, so the metric measures near-trajectory interpolation rather than spatial generalization. We introduce a fair matched-count protocol that isolates this effect: both arms train on the same number of images and differ only in whether the holdout is spread evenly (interpolation) or forms a contiguous spatial sector (extrapolation). Our primary finding is a large, consistent interpolation-extrapolation gap of 3~12dB -- several times the differences typically reported between competing methods. The gap is robust to training noise, is in two cases large enough to flip a method ranking under multi-seed confirmation, and -- crucially -- persists across three representation families, including a non-Gaussian volumetric neural radiance field (NeRF), so it reflects spatial coverage rather than any one representation. Diagnostically, it is dominated by a diffuse/geometry-proxy component and tracks each view's angular distance to its nearest training view, a zero-cost signal that also guides capture planning; loss-side regularization yields only marginal gains. Standard holdouts remain useful for near-trajectory rendering but should not, alone, be read as evidence of spatial generalization. Prior work notes protocol sensitivity; ours is, to our knowledge, the first to combine matched-count paired holdout, cross-representation quantification, and a diagnostic analysis Table 1. We describe a spatial-holdout benchmark toolkit with standardized splits and baselines for 16 scenes, which we are preparing for public release.

---

### Boosting Infrared Small Target Detection via Logit-Domain Contrast and Adaptive Shape Refinement

Authors: Handong Zeng, Zhengeng Yang, Shuai Zhang, Shikai Chen, Hongshan Yu
Published: 2026-07-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2607.01555v1

Infrared small target detection (IRSTD) remains challenging due to tiny target size, low signal-to-noise ratio, severe foreground-background imbalance, and blurred boundaries in complex scenes. Existing methods usually rely on post-activation probability-domain supervision for discrimination, where weak targets and strong clutter may produce saturated and close probabilities, limiting weak-target discrimination. Meanwhile, blurred boundaries and halo-like predictions mainly stem from thermal diffusion, tiny target scale, boundary uncertainty, and insufficient explicit contour constraints. To address these issues, we propose Adaptive-Contrastive SLSIoU (AC-SLSIoU), a plug-and-play discriminative and shape-aware loss for IRSTD. Specifically, a Logit-Domain Margin Constraint (LDMC) is introduced to enlarge the response gap between targets and informative hard negatives in the logit space, thereby enhancing weak-target discrimination. Adaptive Boundary Suppression (ABS) applies scale-aware annular penalties to refine target contours and suppress halo-like overflow responses. In addition, False-Alarm Focal Loss assigns larger weights to high-probability negative samples, further penalizing persistent high-confidence false alarms. Without introducing extra inference overhead, the proposed method can be seamlessly integrated into existing detectors and consistently improves both detection accuracy and shape quality. Extensive experiments and cross-backbone evaluations demonstrate the effectiveness, robustness, and generalization ability of the proposed method for infrared small target detection.

---

### X-LogSMask: Expand Transformer for Graph-Structured Data

Authors: Leyan Li, Rennong Yang, Zhenxing Zhang, Liping Hu
Published: 2026-07-02
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2607.01553v1

Transformers have become general-purpose architectures, but their all-to-all self-attention is poorly matched to graph data, whose interactions are sparse, structured and multi-scale. Existing Graph Transformers address this mismatch through structural encodings, hybrid message-passing modules or learned attention constraints, often introducing additional complexity and limited interpretability. Here we introduce X-LogSMask, an explainable multi-head logarithmic structural mask that injects symmetrically normalized graph topology directly into attention logits. The logarithmic transform converts structural connectivity into a topology-aware gating signal, suppressing unsupported node interactions while preserving feature-dependent attention. By assigning different powers of the normalized adjacency matrix to different attention heads, X-LogSMask gives each head a defined structural radius and supports multi-hop information propagation within a single layer. We further show that a standard Transformer encoder can be interpreted as one-step message passing on a complete graph, motivating X-LogSMask as a topology-constrained alternative to unrestricted self-attention. Across 20 node-, edge- and graph-level benchmarks, Transformers equipped with X-LogSMask achieve state-of-the-art performance on 13 datasets and remain competitive in a lightweight one-layer configuration. These results show that simple, interpretable structural masks can make self-attention an effective graph-learning operator without changing the Transformer architecture. The code is available at https://github.com/LiLeyan-0120/X-LogSMask.

---

### Mechanism and Stability Analysis of Metabolic Closed-Loop Metaheuristics

Authors: Jinliang Xu, Liping Ma
Published: 2026-07-02
Categories: cs.NE, cs.MA
arXiv: https://arxiv.org/abs/2607.01551v2

This paper studies the Metabolic Multi-Agent Optimizer (MMAO) at the framework level rather than at the implementation or benchmark level. The central question is whether the metabolic resource loop of private energy, communal budget, role drift, and lifecycle turnover has a framework-level interpretation beyond narrative metaphor. We introduce a generic MMAO state model that abstracts away domain-specific move operators while retaining the resource bookkeeping that defines the framework. Under mild bounded-gain and bounded-spending assumptions, we establish boundedness and nonnegativity properties for private energy, communal budget, role state, and active population size. We then characterize three endogenous behavioral regimes of the loop: contraction under sustained resource deficit, reinvestment under surplus communal accumulation, and search redistribution under heterogeneous marginal returns across agents or subgroups. The analysis is intentionally conservative. It does not claim global convergence of the full adaptive system, universal superiority over specialist optimizers, or a complete stationary characterization of the resulting process. Instead, it identifies which internal regulation properties are generic consequences of the loop and which remain implementation specific. A compact mechanism-validation package on representative continuous and discrete MMAO realizations provides supporting empirical evidence for this reading, but is not intended to replace a full benchmark study. The resulting contribution is therefore a bounded, regenerative, resource-regulated interpretation of MMAO, rather than a complete proof of all adaptive behaviors of the full algorithm family.

---

### Evolutionary Feature Engineering for Structured Data

Authors: Ege Onur Taga, Yilin Zhuang, M. Emrullah Ildiz, Petros Mol, Abhimanyu Das, Karthik Duraisamy, Samet Oymak
Published: 2026-07-02
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2607.01548v1

Large language models are increasingly used as open-ended search operators in evolutionary optimization. We introduce Evolutionary Feature Engineering (EFE), a framework for using LLM-based evolution to discover preprocessing transformations for structured data. EFE represents transformations as Python programs with a standardized fit/transform interface, allowing them to be inserted directly into existing machine learning pipelines. During evolution, candidate programs are refined using dataset context, summary statistics, and downstream performance feedback on validation set. We instantiate EFE in two settings. For time-series forecasting, EFE-Time learns invertible, dataset-specific normalizations that improve off-the-shelf time-series foundation models. It reduces forecasting errors (MASE, WQL, MAE) 3% or more when averaged across datasets and improvements are as much as 19% on the COVID-Deaths dataset. Notably, these improvements occur with recent TSFMs such as Chronos-2. For tabular prediction, EFE-Tab evolves compact feature programs that add useful interpretable features and remove redundant ones, improving or matching existing LLM-based feature-engineering methods. We found EFE-Tab to be particularly effective on classical decision trees, where small sets of evolved features yield competitive accuracy while preserving interpretability. Overall, EFE demonstrates that LLM-based evolution can improve both accuracy and interpretability when automatically tackling structured data.

---

### BudgetBench: A Budget-Tiered Protocol and Pilot Harness for Memory Strategy Evaluation in Local Large Language Model Agents

Authors: Aditya Karnam Gururaj Rao, Arjun Jaggi
Published: 2026-07-01
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2609.13149v1

For local large language model agents, active context is a scarce resource: memory capacity, prefill latency, cache growth, and service objectives all constrain how many input tokens each call can afford. We present BudgetBench, an active-budget protocol and reference harness that treats the per-call input-token budget as the independent variable when comparing memory strategies. Holding the model, task, sampler, and decoding fixed, it sweeps budgets over 2K, 4K, 8K, 16K, and 32K tokens and records quality, budget utilization, latency, and, as a first-class outcome, budget-violation rates. The core contribution is this reusable measurement surface: a swappable MemoryStrategy contract, explicit budget enforcement, deterministic or versioned graders, prompt-audit metadata, and reproducibility artifacts, released at https://github.com/aviskaar/budgetbench. We substantiate the protocol with pilot studies rather than final rankings. Across a local qwen2.5:1.5b pilot (89 items each on SWE-bench Verified and LongBench v2), a hosted 50-item Qwen3 30B-A3B LongBench replication with exact tokenization, and a 500-item LongMemEval oracle study scored by the official GPT-4o evaluator, the harness exposes budget-compliance failures, non-monotonic quality curves, and operating points that single-budget evaluation hides. The budgeted-versus-full-context direction remains unresolved: the local slice is near-null and the hosted replication favors full context in point estimate. We report results transparently, including that the early pilot's tokenizer approximation undercounts some served-model prompts, so its violation rows are tokenizer-approximation diagnostics, not claim-bearing results; all timings are operational diagnostics. The reusable contribution is the protocol, harness, and failure-reporting discipline needed to scale fixed-budget memory-strategy evaluation.

---

### Optimizing AI Inference Across the Deployment Stack

Authors: Tejinder Singh, John Pflueger, Jeebak Mitra, Robert Lincourt, Mitchell Markow, Bhavesh A. Patel
Published: 2026-07-01
Categories: cs.SE, cs.LG
arXiv: https://arxiv.org/abs/2609.10550v1

AI deployment performance is shaped not by model architecture alone, but by interactions among compression, compiler transformations, and serving policies. Published benchmarks often report latency and throughput under incomparable conditions, limiting their use for deployment decisions. This paper presents a unified analytical treatment of inference optimization across the deployment stack. We introduce a three-layer taxonomy covering model-level techniques such as quantization, pruning, and distillation; compiler transformations such as graph fusion, layout optimization, and kernel autotuning; and system policies such as dynamic batching, admission control, and memory tiering. We formulate deployment as a constrained multi-objective optimization problem over accuracy, latency, throughput, memory footprint, and energy, and analyze a deployment-ranking functional with Pareto monotonicity and scale invariance. Roofline models show how memory-bandwidth hierarchies bound performance across precision regimes, while queuing models explain how service-time changes amplify response time under load. To improve comparability, we propose an evidence protocol that separates measured, derived, and analytical claims; limits numerical comparison to within-paper results; and requires reporting of hardware, software versions, batch semantics, and thermal state. We synthesize evidence from edge platforms, including Jetson AGX Orin and five inference frameworks; data center GPUs, including A100 and H100 with three LLM serving engines; and quantization studies across the Llama-3.1 family. The synthesis shows that deployment outcomes are governed by cross-layer interactions that no single-layer analysis can predict. We conclude with a constraint-aware selection procedure and open problems in compiler-serving co-optimization, cross-hardware performance prediction, and standardized energy reporting.

---

### Physics-informed neural networks by Gradient-Guided Gaussian Adaptive Sampling (3GAS-PINNs)

Authors: Yousen Wang, Wei Zhao
Published: 2026-07-01
Categories: cs.LG, physics.comp-ph, physics.flu-dyn
arXiv: https://arxiv.org/abs/2609.09162v1

Physics-informed neural networks (PINNs) provide a mesh-free framework for solving partial differential equations, yet their performance in nonlinear problems is often limited by slow convergence, gradient imbalance, and insufficient resolution to capture localized intermittent structures such as shock waves(1). These issues arise primarily from the use of fixed weights of loss and uniform collocation point distributions, which cannot adapt to the evolving complexity of the solution field during training. To address these challenges, Gradient-Guided Gaussian Adaptive Sampling Physics-Informed Neural Networks (3GAS-PINNs) is proposed in this paper, which combines uniform probability distribution and Gaussian-smoothed probability distribution derived from the spatial gradients of solution, to maintain global constraint satisfaction as well as concentrating collocation points in regions of high gradient. Thus, intermittency structures like shock wave and solitons can be accurately captured. The method is evaluated on three benchmark nonlinear problems, including one-dimensional forced Burgers equation, Korteweg-de Vries (KdV) equation and nonlinear Schrodinger equation, all of which exhibit steep gradients or strong nonlinearity. In comparison with baseline PINNs, 3GAS-PINNs can effectively promote the physical consistency in intermittent regions. The accuracy of the numerical simulation can be improved by a factor of up to 14.

---

### R$^{2}$Adapter: A Routing and Rewriting Adapter for Efficient Hybrid RAG

Authors: Yucan Guo, Miao Su, Saiping Guan, Long Bai, Zhongni Hou, Zixuan Li, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng
Published: 2026-07-01
Categories: cs.CL, cs.IR
arXiv: https://arxiv.org/abs/2609.02894v1

Retrieval-Augmented Generation (RAG) has become a prevailing paradigm for enhancing Large Language Models (LLMs) with non-parametric knowledge. Vanilla RAG efficiently handles simple queries but struggles with relational or multi-hop reasoning. Graph-based RAG alleviates this issue but incurs higher inference complexity and latency. In practice, user queries can differ significantly in their complexity, rendering a fixed RAG strategy suboptimal. However, existing hybrid text-graph RAG methods typically rely on heuristic and LLM-based routing, resulting in unnecessary overhead and strong dependence on the underlying LLM. To address these challenges, we propose R$^{2}$Adapter, a lightweight plug-in Routing and Rewriting Adapter designed to allocate queries between vanilla and graph-based RAG dynamically. By routing only the queries that genuinely benefit from graph-based reasoning, R$^{2}$Adapter reduces unnecessary graph retrieval overhead. Additionally, uncertain graph-routed queries are rewritten to better expose their multi-hop reasoning requirements, improving retrieval quality without additional supervision. Extensive experiments on three multi-hop QA benchmarks demonstrate that R$^{2}$Adapter reduces graph-based RAG usage by up to 59% while maintaining comparable answer accuracy. This adapter is model-agnostic and can be seamlessly integrated into diverse vanilla and graph-based RAG pipelines, providing an efficient and adaptive solution for hybrid RAG systems.

---

### Probe Generalization as Subspace Selection for OOD Deception Detection

Authors: Daniel Yoo, Adrians Skapars
Published: 2026-07-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2609.02893v1

Linear probes can be used to detect behaviors and concepts inside language model activations, but may fail to transfer to out-of-distribution examples. When studying the generalization performance of Llama-3.1-8B-Instruct probes over 3 held-out deception detection datasets, we find that projecting inputs onto a small subset of principal components (PCs) from the training distribution of activations enables cross-domain transfer that nearly matches the performance of probes trained directly on the test distribution. Furthermore, we find that PC interpretations can be used to find a subset of those transferable PCs. By using an LLM judge to score each PC on whether its most/ least activating examples imply a transferable deception direction, then probing on the highest-scoring PCs, we close the baseline-to-oracle gap by 78% on Insider Trading Report and by 25% on Sandbagging. The directions a source probe weights heavily appear to encode source-specific surface features, while the directions that actually transfer appear to encode the same contrast more abstractly, in a way natural language descriptions can capture. Broadly, our results suggest that the OOD robustness of probes is largely determined by subspace selection.

---

### Counterexamples as Feedback for Agent Self-Correction

Authors: Sidhesh Badrinarayan, Adithya Parthasarathy
Published: 2026-07-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2609.02892v1

Single-turn code-generation metrics understate a central property of deployed agents: whether they can repair a wrong artifact after receiving concrete feedback. This paper presents A-CEGIS, a lightweight framework that uses counterexamples as feedback for evaluating multi-turn refinement in natural-language-to-regex synthesis. An agent proposes a regex, a deterministic oracle checks it under full-match semantics, and compact false-positive or false-negative witnesses guide the next turn. On 30 NL-RX-Turk tasks, diagnostic counterexample feedback solves 90\% of tasks within a four-turn ablation budget, compared with 17% for zero-shot generation, 27% for generic self-correction, and 23% for error-only feedback. In a full diagnostic run with hardening, all tasks are solved on the hidden set by the final turn, with mean time-to-success of 2.7 turns and robust success of 77% after targeted probing. These results show that A-CEGIS measures how efficiently an agent improves across turns while adding a practical robustness check beyond the original held-out cases.

---

### SciReC: Diagnostic Evaluation of Multimodal, Multi-Turn Relational Reasoning with Adaptive Interaction

Authors: Nilay Yilmaz, Naga Sai Abhiram Kusumba, Stella Wenxing Liu, Yezhou Yang
Published: 2026-07-01
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2608.27461v1

Relational reasoning requires the process of perceptual understanding, comparing, and integrating the underlying relationships between concepts. This ability consists of multiple categories, such as analogical, structural, and cause-effect, each capturing a different aspect of higher-order understanding. To examine the performance of multimodal large language models (MLLM) on these relational inference tasks, we developed SciReC, a model-adaptive multimodal academic dialog benchmark. As the relational reasoning process involves multiple representations and various factors (visual understanding, exhibiting knowledge, and memory recall), we propose DMRA, a deficit-based diagnostic framework that quantifies the contribution of these components to identify the primary cause of unsuccessful cases. Claude 4.6 achieved the best performance on the overall relational score with 73\%, followed by GPT 5.4 with 68\%. Performance trends indicate that open-source models achieve their lowest scores on spatial relations, while proprietary models struggle more with hierarchical and sequential relations. Across domains, model performance is lowest on Astronomy and highest on Psychology. The results of DMRA reveal that relational reasoning is the primary source of error across all models, followed by memory limitations.

---

### Accelerating LLM Inference via Vector Index Based Output Embeddings

Authors: Martin Loretz, Sepp Hochreiter
Published: 2026-07-01
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2608.27460v1

Large output embedding matrices create a significant memory bandwidth bottleneck during autoregressive decoding, especially for compact LLMs with large multilingual vocabularies. We reformulate the output projection followed by top-k token selection as a maximum inner product search over token embeddings and replace the dense vocabulary projection with an HNSW-based vector index. The resulting output head retrieves only a small candidate set of high-scoring tokens and can be integrated into existing decoding pipelines by scattering retrieved logits into a sparse full-vocabulary tensor. On CPU inference with Gemma 3, Llama 3.2, and Qwen 3 models, our method substantially accelerates the output projection and improves end-to-end batch-size-one decoding throughput by up to 82% for Gemma 3 270M, while preserving generation quality under AlpacaEval evaluation. These results suggest approximate retrieval is a practical alternative to dense output projections in latency-sensitive small-batch decoding.

---

### Explainable Artificial Intelligence for Customer Churn Prediction in Telecommunications: A Framework for CRM Integration

Authors: Sandeep Gaddamwar
Published: 2026-07-01
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2608.26151v1

Subscriber attrition is a costly, persistent challenge for telecommunications providers, with monthly churn of roughly 1.9% in mature markets eroding billions in revenue annually. Predictive models can flag at-risk customers accurately, yet they are routinely excluded from frontline CRM workflows because high-performing ensemble and non-linear architectures are opaque: a retention specialist cannot design a personalised intervention from a probability score alone, without knowing why a subscriber is at risk. This paper addresses that gap. We benchmark four classifiers--Logistic Regression, Random Forest, XGBoost, and LightGBM--on the IBM Telco Customer Churn benchmark (7,043 records; 19 features; 26.5% churn, balanced to 50% via SMOTE on the training partition only). Logistic Regression attains the strongest AUC-ROC (0.8411) and LightGBM the highest accuracy (78.42%); all four fall within a 0.011 AUC band (0.831--0.841), and 5-fold cross-validation confirms the leading models are effectively tied. Explanations are delivered at two granularities: a global SHAP ranking identifying tenure, total charges, and month-to-month contract as the dominant churn signals, and instance-level SHAP and LIME decompositions that expose the drivers behind each prediction. Building on these outputs, we introduce a four-layer CRM integration architecture that converts risk scores and attribution vectors into tiered segmentation, maps top features to structured retention-action templates, and routes campaign outcomes into a retraining feedback loop. Targeting the highest-risk quintile is projected to cut overall churn by 3.3--5.3 percentage points, preserving an estimated $199K--$319K per campaign cycle.

---

### Leveraging Large Language Models for Systematic Literature Review of Disease Spread Models

Authors: Orhan Yagizer Cinar, Timur Emre Ozkose, Emma Von Hoene, Amira Roess, Taylor Anderson, Hamdi Kavak
Published: 2026-07-01
Categories: cs.AI, cs.DL, cs.IR
arXiv: https://arxiv.org/abs/2608.26150v1

Recent advancements in Large Language Models (LLMs) have created new opportunities to streamline and potentially automate many research processes, including systematic literature reviews (SLRs). This study reports an LLM pipeline development for extracting model-relevant information from 536 peer-reviewed agent-based modeling papers. We compare the results with those of a human-conducted SLR. Our results show paper-level accuracies of approximately 77.95% for GPT-4.1 and 81.67% for GPT-5.0. Field-level accuracy ranges from 32.40% to 100.00%, with more complex or subjective fields performing less reliably. Importantly, we find that agreement between LLMs is a potential indicator of output quality: low agreement may signal hallucinations, whereas high agreement combined with low accuracy may point to noise or errors in the human dataset. Overall, our study provides practical insights into prompt development and highlights both the potential and limitations of using LLMs for full-scale SLRs in the modeling and simulation domain.

---

### Methodological and Conceptual Framework for 5D Multi-Table Analysis: A Unified Approach for Complex Data Reuse

Authors: Edouard Lansiaux, Hugo Kazzi, Aurélien Loison, Slim Hammadi, Emmanuel Chazard
Published: 2026-07-01
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2608.26149v1

Multi-table learning remains a major challenge in machine learning for healthcare and other complex information systems. Relational data combine several sources of complexity, including large data volume, high-dimensional variables, high-cardinality categorical features, complex inter-table dependencies, and repeated temporal observations. We introduce the Relational Hypergraph Transformer (RHT), a unified architecture that represents relational databases as hypergraphs, learns pentadimensional embeddings (PentE), and performs sparse relational attention with complexity proportional to the average relational degree rather than the square of the number of entities. We formally define the architecture, derive the complexity of its attention mechanism, and provide an open-source reference implementation. We evaluate RHT on the public Synthea synthetic electronic health record dataset using multi-label prediction of SNOMED CT condition codes per encounter, a task characterized by high categorical cardinality and long-tailed label distributions. Comparisons with tabular, relational, and temporal graph baselines show that RHT produces more semantically coherent embeddings while remaining computationally scalable. In this benchmark, the highest rare-code recall is achieved by XGBoost, whereas RHT attains the strongest embedding semantic coherence. We also report ablation studies quantifying the contribution of each architectural component. Clinical validation on MIMIC-IV is planned following PhysioNet credentialing. Source code and experimental protocols are provided in the accompanying repository.

---

### LitReview Arena: Evaluating Literature Review Agents with Battle-Style Peer Review Platform

Authors: Ruotong Zhao, Zhiyu Chen, Xurui Liu, Haidong Xue, Dong Liang, Jigao Fu, Wu YanBiao, Yuanyi Zhen, Fengli Xu, Yong Li
Published: 2026-07-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2608.21374v2

Literature reviews are essential to scientific progress, but rigorously evaluating automatically generated reviews remains difficult because many aspects of research utility depend on expert judgment rather than reference-overlap metrics. We introduce LitReview Arena, a battle-style evaluation platform with a structured protocol tailored to literature review quality: domain experts with AI paper-writing experience compare anonymized drafts, are matched to topics within their expertise, and provide dimension-wise outcomes over five literature-review-specific criteria. From this protocol, we collect approximately 3k expert judgments, each containing five dimension-wise outcomes, and show that even the strongest current systems win only 23.0% of decisive matches against human drafts on overall utility, while agentic LLMs such as Sonar Deep Research substantially outperform base language models by over 60%. We further find that existing LLM-as-a-judge methods are substantially misaligned with human experts (Spearman's rho=0.467), especially on synthesis-heavy criteria such as paper structure and research suggestions. Using the collected preference data, we provide an expert-calibrated evaluator, LitJudge, which improves alignment to Spearman's rho=0.78, comparable to inter-expert consistency; code and data are publicly available at https://github.com/VanellopeAsher/LitReview-Arena.

---

### Nexus: Depth-Adaptive KV-Cache Splicing and Retrieval-Decoupled Tool Routing for Agentic LLMs on Unified Memory

Authors: Mustafa Arslan
Published: 2026-07-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2608.20397v1

Agentic large language models (LLMs) on the Model Context Protocol (MCP) re-encode verbose tool schemas every turn, so prefill - quadratic in sequence length - dominates time-to-first-token (TTFT) as the tool registry grows. Nexus's primary lever is to decouple routing from the schema-prefill cost: an INT8 semantic lookaside buffer (SLB) with a calibrated cross-encoder margin gate selects tools by retrieval, and arguments are generated over a compressed textual signature (median 19 tokens) rather than over spliced key/value (KV) cache. This path is depth-independent: routing accuracy stays near 89% as the registry scales to 250 tools - where a concatenate-all-schemas baseline overflows the context window entirely - and it reaches a first-argument token 1.66x sooner than a full-schema re-prefill at a ~80% main-context token saving. As a secondary, bounded lever we transplant a compiled schema KV block directly into the live context. This is fundamentally limited by rotary position embedding (RoPE) phase drift: an anchored splice is output-exact, but off-anchor placement corrupts attention, so beyond a threshold P=256 Nexus repairs the seam with a depth-adaptive suffix redecode that escalates to a full re-prefill. The resulting never-regress property is a guarantee on output fidelity (top-1 agreement, D_KL approx. 0) - not on latency, which can dip to 0.98x before converging to parity - alongside a 1.1-1.7x TTFT speedup at moderate depth that narrows to parity at deep context. Two negative results bound the design: the off-anchor RoPE fidelity boundary, and the failure of a reference-free drift gate to predict drift (Spearman rho = 0.193). All measurements are from one model tuple (Qwen2.5-14B-Instruct Q4_K_M) on Apple-silicon unified memory; the qualitative boundaries generalize, while the quantitative envelope is tuple-specific.

---

### Self-Supervised Speech Representations Track Spoken Language Convergence to Adult Models in Infants and Children Who Are Deaf/Hard-of-Hearing

Authors: L. Choy, A. S. Khan, S. Patrizi, D. Ye, J. Gross, M. Cychosz
Published: 2026-07-01
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2608.20396v1

Language development is characterized by a gradual convergence of children's speech toward adult patterns. Measuring this process has traditionally required detailed transcription and language-specific expertise, limiting scalability across languages and populations. Here, we use speech embeddings to capture this convergence directly from the acoustic signal in longform, child-centered recordings, taken as children go about their daily lives. Using HuBERT-BASE, we extracted embeddings from speech vocalizations of children who are deaf/hard-of-hearing and their female adult caregivers ($>$925 hrs. observation). Embedding distance between children and caregivers decreased with hearing age, controlling for pitch and vocalization length, indicating, as expected, that children's speech patterns converge to caregivers over development. This single distance metric likewise related to multiple standardized measures of speech and language from infancy through preschoolhood. These results suggest a path toward scalable, language-neutral assessment of spoken language development from children's everyday lives.

---

### A Factorial Ablation of a Speech-to-SFT Pipeline: Differential Effects on Data Quality and Downstream Transfer

Authors: Wonsup Shin, Jingu Kim
Published: 2026-07-01
Categories: cs.SD, cs.CL
arXiv: https://arxiv.org/abs/2608.20394v1

Industry pipelines that turn speech into supervised fine-tuning (SFT) data via multi-stage refinement are increasingly adopted but, to our knowledge, have not been publicly ablated stage-by-stage, leaving each stage's marginal value unknown. We design a production-ready speech-to-SFT pipeline in which transcript refinement (Phase 0) and SFT data quality refinement (Phase 2) are independently toggleable, yielding a 2x2 factorial design. For each condition, we generate QA-form SFT data from Korean medical and finance conference recordings and fine-tune 9 models (5 LLM families, 2.4B-70B); we evaluate with four cross-provider LLM judges, a blind six-expert human evaluation, and 3 downstream MCQA benchmarks. Our central finding: under a fixed, standard SFT recipe, improvements in QA data quality do not transfer uniformly into downstream MCQA gains. 4-judge quality rises consistently, yet the cross-model mean MCQA gain is not significant; positive transfer concentrates on family-domain aligned pairs. This differential pattern is consistent with a format mismatch: Phase 2 shifts SFT-data composition toward explanatory items, while MCQA primarily probes factoid recall. All six human raters report higher full-pipeline quality, confirming the LLM-judge direction. An STT-engine swap to Whisper-medium confirms pipeline robustness. A non-hallucination audit shows the two frontier LLMs admit unknown on approximately 8% of QA on average; we release samples, prompts, code, and all SFT checkpoints.

---

### Knowledge-Graph-Gated Defactualization for Style-Controllable and Fact-Preserving Generation in Agentic Conversational AI

Authors: Tanmay Kumar Shrivastava, Darsh Rohit Nandu, Rajesh Kumar Mundotiya
Published: 2026-07-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2608.20393v1

Agentic large language models (LLMs) deployed in fact-sensitive applications such as customer support must simultaneously preserve factual correctness and generate responses in a controllable stylistic register. Activation steering enables fine-tuning-free style control by perturbing hidden representations, but it lacks an explicit mechanism for distinguishing verifiable facts from stylistic content, leading to semantic leakage. We address this challenge through \emph{Defactualize-Steer-Rehydrate} (DSR), a knowledge-engineering framework that integrates a typed, salience-weighted knowledge graph (KG) with activation steering. DSR extracts salient entities using a layered regex or NER or lexical-classifier pipeline, replaces them with typed placeholders prior to steering, and deterministically restores verified values through salience-guided rehydration after generation. DSR is evaluated across six LLaMA-family models (1B--13B parameters) on 600 A2A-generated customer-support cases (1,200 generations), with a dedicated KG ablation study. DSR significantly increases verified-entity recovery relative to a steering-only baseline (Cohen's $d=0.225$, $p_{\text{Bonf}}=1.0\times10^{-4}$), though the absolute recovery rate remains modest, while preserving effective style control across diverse model families. Layer-wise separability and steering-strength diagnostics further show previously unexplored interactions between representation-level steering and factual grounding. hese results demonstrate that explicit knowledge engineering can systematically enhance trustworthy, controllable, and reproducible generative AI without requiring model fine-tuning. Code, cached steering vectors, and evaluation scripts are publicly released to support reproducibility.\footnote{https://github.com/Tanmay-IITDSAI/KG-Gated-Defactualization}

---

### You Are What You Prompt: Prompt Quality, Domain Shift, and Uncertainty in Agrifood Vision-Language Models

Authors: Andrea Morales-Garzón, Salvador López-Joya, Miguel López-Pérez, Maria J. Martin-Bautista
Published: 2026-07-01
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2608.18116v1

Vision-language models enable zero-shot classification through natural language prompts, but performance is sensitive to prompt formulation, especially in specialized domains. Zero-shot Prompt Ensembling (ZPE) addresses this by weighting prompts by discriminative signal, yet its behavior under domain shift remains unexplored. We evaluate ZPE in the agrifood domain using CLIP and SigLIP across four datasets and four prompt pools, spanning in-distribution (ID) food and out-of-distribution agricultural benchmarks. ZPE provides limited benefit under ID conditions but substantially improves performance and calibration under domain shift, where domain-specific pools of 51-52 prompts consistently outperform generic pools of 247-426. Lexical analysis shows that ZPE acts as an unsupervised domain-alignment detector without label access. We further introduce PID (Prompt-based Inconsistency Detection), which repurposes prompt disagreement as epistemic uncertainty, improving failure detection under severe domain shift where standard confidence measures collapse.

---

### Demo: Real-time Generative Multicasting with On-Device Intent-aware Semantic Decomposition

Authors: Xinkai Liu, Mahdi Boloursaz Mashhadi, Yi Ma, Rahim Tafazolli
Published: 2026-07-01
Categories: cs.NI, cs.LG, cs.MM
arXiv: https://arxiv.org/abs/2608.14600v1

We present a demonstration for generative multicasting with on-device, intent-aware semantic decomposition. At the transmitter, DNN-based segmentation extracts a semantic map from the source video, decomposing it into multiple sub-signal classes based on multi-user receiver intents. The transmitter broadcasts the semantic map to all users over shared wireless/network resources, thereby utilizing orthogonal resources only to transmit the sub-signal classes intended for each user. Users partially reconstruct and partially synthesize the signal by combining the received intended classes with non-intended classes locally synthesized by a generative model from the semantic map. We derive the rate-distortion/perception curves for reconstruction/synthesis with the generative model, to adaptively set compression rates for the semantic map and intended classes. Generative multicasting significantly reduces the wireless/network resources required for existing/emerging multimedia multicasting applications. The system is real-time on a Google Coral Edge TPU with 4 TOPS (int8). This is the first demonstration of generative multicasting representing a substantial advancement in on-device generative SemCom.

---

### Think in Latent, Explain in Language: Self-Explainable Latent Reasoning

Authors: Dayuan Zhao, Shengcao Cao, Yu-Xiong Wang, Liang-Yan Gui
Published: 2026-07-01
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2608.13570v1

Latent reasoning has emerged as a powerful alternative to text-based Chain-of-Thought (CoT), offering significant gains in computational efficiency by compressing verbose reasoning into compact embeddings. However, compressing reasoning into the latent space renders the thinking opaque, hindering its interpretability. Current methods present a stark trade-off: they either function as unexplainable ''black boxes'' (e.g., Coconut), where the latent reasoning is not human-readable, or rely on separate post-hoc decoders for explainability (e.g., Heima), introducing architectural overhead and decoupling the explanation from the actual reasoning process. In this work, we present a unified framework for Self-Explainable Latent Reasoning (SELR) that trains a single model to perform efficient and inherently explainable latent reasoning. Our core contribution is a novel multi-task training objective that optimizes for two goals simultaneously: (1) an Answer Loss that optimizes the latent reasoning trajectory to produce accurate final answers, and (2) a CoT Loss that explicitly trains the same model to decode its own latent representations back into human-understandable reasoning steps. This design ensures that generated latent representations are both task-effective and semantically interpretable, eliminating the need for external decoders. We validate the effectiveness of SELR on both Large Language Models (LLMs) and Vision-Language Models (VLMs), demonstrating that SELR achieves superior token efficiency and accuracy compared to baselines, while uniquely providing self-contained explainability without auxiliary models. Project page is available at https://jasondayuan.github.io/SELR/.

---

### Navigation Alone Is Not Enough: Evaluating Explanatory Assistive UI Agents

Authors: Santosh Patapati
Published: 2026-07-01
Categories: cs.HC, cs.AI
arXiv: https://arxiv.org/abs/2608.09944v1

Modern web interfaces are increasingly difficult to use with screen readers, particularly when pages update dynamically or hide important structure behind visual layout. Recent UI agents can act on such interfaces; however, for assistive agents to be truly useful, they must behave as collaborators that keep users informed and in control, rather than as tools that simply take actions on users' behalf. Most existing benchmarks judge systems primarily by task completion, without assessing how well they explain their actions or support user oversight. We introduce NeXUI, a benchmark for assistive agents that must navigate interfaces while explaining each step in clear language for nonvisual use. NeXUI pairs realistic user goals with instrumented interface states, enabling agents to reason from both visual context and structural information. Its evaluation measures safety, efficiency, and task success, while also checking whether explanations are grounded in the interface state. In our experiments, we find that NeXUI remains challenging even for state-of-the-art foundation models, with % Gemini-3.5-Flash achieving only a 44\% success rate and poor explanation scores, making it a useful foundation for future research and development. By focusing on navigation, explanation, and user control, NeXUI provides a clearer way to study agents that can support blind and visually impaired users in modern computing environments.

---

### EweAcT: Ewe behaviour aligned to accelerometer data for activity monitoring in extensive grazing systems

Authors: Lucile Riaboff, Ny Aina Andriamampandry, Jean-François Bompa, Mathias Aletru, Christian Durand, Sébastien Douls, Gaëtan Bonnafe, Morgane Costes-Thiré, Guillaume Delosières, Jean- Marc Mongrelet, Enzo Niro, Némuel Tadi et al.
Published: 2026-07-01
Categories: cs.HC, cs.LG
arXiv: https://arxiv.org/abs/2608.09943v1

Monitoring livestock behaviour under extensive conditions would provide valuable insights to assess animal adaption to environmental perturbations in agroecological systems (e.g., heat waves, parasitism, predator attacks). Animal behaviour can be monitored using accelerometer data collected from neck-collars combined with artificial intelligence models. However, large amounts of accelerometer data aligned with annotated behaviours are necessary to develop accurate models of behaviour prediction. In particular, developing reliable models for extensive systems requires data collected across a wide range of representative conditions. The dataset includes 79 hours of tri-axial accelerometer data aligned with behaviours manually annotated from video recordings for 120 Romane ewes born between 2021 and 2024. The ewes were derived from two divergent genetic lines after three and four generations of selection started 10 years ago: low and high social attractiveness, noted S-and S+, and low and high tolerance towards humans, noted H-and H+. They were reared under the extensive system applied to the Experimental Unit of La Fage (UEF, INRAE, Saint-Jean-et-saint Paul, Aveyron) where 250 sheep were reared exclusively outdoors on 280 hectares of rangeland in southern France. First batch of data was collected on March, June and July 2024 at the UEF under a range of extensive conditions, including sloping pastures and heat-wave periods. Ewes were equipped with accelerometer neck-collars specifically designed for young sheep on pasture. They were grouped on experimental paddocks for 4 to 8 hours and provided with fresh grass and ad libitum access to water. The animals were simultaneously video-recorded using an elevated CCTV camera. Behaviour annotation was carried out using Behavioral Observation Research Interactive Software focusing on the main behaviours on pasture: Grazing, Ruminating, Resting, Moving, and ''Other'', grouping all remaining activities. Annotations and correspon…

---

### Automatic Detection of Deaths from Social Networking Sites

Authors: Nuhu Ibrahim, Riza Batista-Navarro
Published: 2026-07-01
Categories: cs.SI, cs.AI
arXiv: https://arxiv.org/abs/2608.05183v1

This dissertation analysed and discussed the differences in linguistic characteristics between pre-mortem and post-mortem social media content, and reported machine learning (ML) classifiers that achieved high performance in automatically detecting deaths of social networking site users from posts associated with their profiles. A new dataset was developed using Wikidata and Twitter. ML models, both traditional (RF, KNN, LR, and SVM) and deep learning (BiLSTM, CNN, and the state-of-the-art BERT), were trained on features extracted using TF-IDF and pre-trained embeddings (Glove, Word2Vec, and FastText) to classify post-mortem content from its pre-mortem counterpart. The results showed that RF outperformed all other traditional ML models; BiLSTM outperformed CNN; TF-IDF consistently outperformed pre-trained word embeddings for the traditional models; Word2Vec consistently outperformed Glove and FastText for the deep learning models; and BERT outperformed all other models. It was found that although pre-mortem and post-mortem tweets express similar levels of positive sentiment, post-mortem tweets exhibit higher negative sentiment, whereas pre-mortem tweets exhibit higher neutral sentiment. Feelings suggesting negativity (sad, angry, surprise, and fear) are more dominant in post-mortem tweets, while happy is more dominant in pre-mortem tweets. It was also found that words, personal pronouns, verbs, family words, religious words, death words, and swear words occur more frequently in post-mortem tweets, whereas impersonal pronouns and informal words occur more frequently in pre-mortem tweets. Additionally, analytical thinking is expressed more in post-mortem than pre-mortem conversations. This experiment's significant contribution is the successful development of an exceptionally high-performing technique for automatically detecting user deaths on social networking sites.

---

### JudgeArena: A Unified Framework for Reproducible LLM-Judge Evaluation

Authors: Erlis Lushtaku, Bora Kargi, Ali Elganzory, Fabio Ferreira, Alejandro R. Salamanca, Julia Kreutzer, David Salinas
Published: 2026-07-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2608.02620v1

LLM-as-a-judge evaluation has become a dominant paradigm for ranking language models, yet the ecosystem remains fragmented: most benchmarks ship their own code base, hardcode a specific closed-model judge, and support a single evaluation protocol. This fragmentation makes it difficult to study how design choices--the benchmark, the judge model, the prompt, the inference backend--affect the conclusions we draw about model quality. We introduce JudgeArena, an open-source framework that unifies major LLM-judge benchmarks (AlpacaEval, Arena-Hard, MT-Bench, and m-Arena-Hard) under a single interface with swappable judges and comprehensive metadata logging for increased transparency in reporting and reproducibility. It enables systematic studies of judge choices, as any model accessible via vLLM, llama.cpp, or OpenRouter can serve as both candidate and judge. Furthermore, JudgeArena ships with tuned judge configurations for open models that match or outperform closed-model judges, validated on human preference datasets in both English and multilingual settings, reducing the reliance on opaque closed models. Finally, by combining existing human annotations with LLM-judge evaluations of a target model, JudgeArena can simulate LMArena Elo scores with high accuracy offering a practical, open, and low-cost alternative to large-scale human annotation campaigns.

---

### Uncertainty-Aware Simulation-Based Inference for Operations Research with Large Language Models

Authors: Liang Guo, Lin Shaochong, Shen Zuo-Jun Max, Zhang Kun
Published: 2026-07-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2608.00019v1

Deploying large language models (LLMs) for operations research (OR) tasks remains challenging because correctness depends on a coherent modeling process, not merely a correct final answer. Standard autoregressive generation operates on a myopic policy, which sometimes fails to anticipate whether a partial formulation can be validly extended into a globally consistent optimization model. Consequently, locally plausible steps may propagate into catastrophic downstream formulation or solver code errors. To address this, we propose an uncertainty-aware, training-free inference framework for OR mathematical modeling. Without updating model parameters, our method evaluates intermediate candidate steps using short lookahead simulations to quantify downstream predictive uncertainty or probability concentration. Candidates that demonstrate a higher likelihood of yielding coherent mathematical formulations are then dynamically selected via importance resampling. Empirical evaluations across multiple OR benchmarks (including NL4OPT, MAMO, and IndustryOR) demonstrate that our framework consistently outperforms both standard and low-temperature baselines, establishing an efficient, training-free paradigm for reliable OR formulation generation.

---

### Building Fast, Evaluating Slow: Pipeline Choices Dominate Autointerpretability Score Variance

Authors: Sinie van der Ben, Neele Roch, Anna Hedström, Mennatallah El-Assady
Published: 2026-07-01
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2607.19386v1

Cross-paper comparison of sparse autoencoder (SAE) interpretability often relies on autointerpretability scores. In this evaluation pipeline, a language model (LM) explains each feature, and another LM scores the explanation. For these comparisons to be meaningful, scores must reflect stable properties of the features rather than confounding aspects of the evaluation pipeline. Through systematic experiments across four metrics (simulation, detection, fuzzing, purity), two models (Pythia-160M, Apertus-8B), and four axes of methodological variation, we show that this assumption does not hold. Specifically, we find that R1) methodological variance collectively exceeds architectural variance across all metrics and tested models; R2) each metric exhibits a distinct instability profile, with detection being the most stable and fuzzing unreliable across all conditions; R3) top-k feature rankings do not stay consistent across corpus and draw conditions, masking per-feature instability behind stable mean scores; a failure that cannot be detected by monitoring explanation similarity alone. These findings suggest that cross-paper comparisons based on autointerpretability scores may reflect pipeline differences rather than architectural differences, with implications for the ongoing debate on SAE utility. More broadly, unreliable evaluation slows progress in interpretability research at a time when reliable tools for understanding AI systems are needed. To support evaluation, we contribute a variance decomposition approach, a Stability Check, and a Minimum Reporting Checklist.

---

### STN-TGAT: Top-K Portfolio Construction via Prior-Guided Graph Attention with Learnable Soft-Threshold Sparsification

Authors: Haoran Guo, Yutong Lu, Li Zhang
Published: 2026-07-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2607.19385v1

This paper tackles the problem of stock ranking and portfolio construction under realistic investment settings by jointly modeling temporal dynamics and cross-sectional dependencies. We propose the Soft-Threshold NMI-prior Transformer Graph Attention Network (STN-TGAT), which integrates a temporal Transformer with a Graph Attention Network to capture long-horizon sequential patterns and dynamic inter-stock relationships. An NMI-based prior graph combined with a soft-threshold sparsification mechanism enhances structural robustness by mitigating noisy correlations while preserving informative connections. The portfolio formation process incorporates practical considerations, including Top-5 selection within the Top-50 $S\&P$ 500 constituents, explicit weight allocation, and transaction cost adjustment, thereby aligning the evaluation with real-world trading conditions. Empirical results on real-world data demonstrate that STN-TGAT consistently outperforms benchmark models from predictive accuracy and investment profitability measured by portfolio returns. These findings suggest that combining decision-aligned training with adaptive relational modeling provides a coherent and practically effective framework for data-driven portfolio construction.

---

### SUM: Unified Geometric Surgery on Spatio-Temporal Adaptation Vectors for Federated Class Incremental Learning

Authors: Jaeik Kim, Jaeyoung Do
Published: 2026-07-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2607.19384v1

Real-world intelligent systems often require both distributed collaboration across data-isolated clients and continual adaptation to evolving tasks. This setting naturally gives rise to Federated Class Incremental Learning (FCIL), which combines Federated Learning (FL) and Continual Learning (CL). However, their combination introduces two coupled sources of interference: spatial interference from heterogeneous clients and temporal interference from sequential tasks, jointly leading to Spatial-Temporal Catastrophic Forgetting (ST-CF). Existing approaches typically address spatial and temporal interference with separate mechanisms, often incurring additional client-side computation or communication, while leaving directional interactions among updates during aggregation unregulated. In this paper, we reinterpret FCIL as a unified multi-task learning problem, where both client and task updates are represented as adaptation vectors in a shared parameter space. Based on this view, we propose Unified Geometric Surgery on Spatio-Temporal Adaptation Vectors (SUM), a purely server-side framework that performs geometric surgery on adaptation vectors during aggregation. Spatial SUM mitigates client-level interference within each round, while causal online temporal SUM removes cross-task interference over time without additional client-side computation, communication, or memory beyond standard federated training. Empirically, SUM achieves up to 22% improvement over prior FCIL methods across diverse vision and language benchmarks while remaining robust to unreliable clients and maintaining computational efficiency.

---

### Trend strength predicts when generative foundation models win: a power-controlled benchmark, a mechanism, and an actionable selection rule

Authors: Ahmed Cherif
Published: 2026-07-01
Categories: stat.AP, cs.LG
arXiv: https://arxiv.org/abs/2607.19383v1

Pretrained generative foundation models cast forecasting as conditional generation from a learned predictive distribution and forecast unseen series zero-shot. We establish three results that turn their reported success into an actionable, mechanistic understanding. First (a positive benchmark result): on a power-controlled study of 1728 rolling-origin forecasts over 36 series from 19 datasets spanning the full range of STL trend strength (F_T in (0.17, 1.00)), a zero-shot Chronos model significantly outperforms four strong classical baselines -- drift, seasonal-naive, Theta, and additive Holt-Winters/ETS -- with the best mean MASE (1.187 vs. Theta 1.337, ETS 1.656; Friedman chi^2 = 46.08, p = 8.75e-09; Holm-corrected Wilcoxon p <= 0.015 against every baseline; a Nemenyi critical difference separating it from the classical pack). Second (a novel, quantified mechanism): a controlled synthetic experiment with a known trend-generating process shows why -- and reveals that the win does not come from better trend extrapolation. When the true trend is linear, damped, or exponential, additive ETS tracks the slope (slope-tracking ratio 1.02, 1.34, 0.98) whereas Chronos systematically under-extrapolates, behaving as a trend-shrinkage estimator (ratio 0.80, 0.49, 0.36). Third (an actionable selection rule): because the advantage is a shrinkage effect, it is predictable from trend strength alone -- the generative model wins 78% of low-trend series but only 44% of high-trend ones, and its edge over ETS is significant on the low-trend stratum (0.982 vs. 1.671, p = 0.002) yet a tie on the high-trend stratum (p = 0.18). Trend strength, computable before forecasting from the training context alone, is therefore a practical a-priori indicator of when to deploy a foundation model. We additionally document a calibration shortfall (80% intervals cover 0.77).

---

### Challenges of Explainability in Continual Learning for Time Series Forecasting

Authors: Quentin Besnard, Emmanuel Doumard, Nicolas Labroche, Nicolas Ragot, Nicolas Ringuet
Published: 2026-07-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2607.19382v1

Deep learning models have shown strong potential for time series forecasting, yet their deployment in real-world environmental monitoring remains challenging due to non-stationary dynamics and limited explainability. In this work, we investigate explainability as a central tool for understanding continual learning in adaptive time series forecasting, with Experience Replay strategies. We study neural forecasting architectures such as PatchMixer, PatchTST and DLinear, augmented with attention-based sampling mechanisms to support model adaptation over time. Explainability is leveraged through attention rollout and gradient-based attribution methods (Grad-CAM) to analyze both predictive behavior and sampling strategies within a continual learning framework. Experiments conducted on real-world piezometric time series exhibiting heterogeneous patterns and regime shifts show that analyzing model and sampling behaviors provides valuable insights into the dynamics of the continual learning framework. Beyond predictive performance, our results highlight the challenges and opportunities of using explainability to understand continual learning behaviors, revealing how attribution patterns evolve over time and how they can inform data selection and adaptation strategies in non-stationary forecasting scenarios.

---

### Air Quality Arena: A Large-Scale Multi-Region Ground Monitoring Dataset and Benchmark for Air Quality Forecasting with Time-Series Foundation Models

Authors: Rishi Bharadwaj, Manik Gupta, Pandarasamy Arjunan
Published: 2026-07-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2607.19381v1

Air pollution causes an estimated 7.9 million premature deaths annually, making accurate forecasting a critical public health priority. Machine learning is increasingly being applied to forecast air pollution levels, yet existing benchmarks remain narrow in both geographic scope and pollutant coverage, and fail to evaluate the latest generation of time series foundation models (TSFMs) on real world, large scale data. We present Air Quality Arena (AQA), a large scale multi-country and multi-pollutant dataset (AQA-Data) and benchmark (AQA-Bench) to address this gap. AQA covers 6 major pollutants over a three year period across 7 diverse countries and 4 continents, with more than 14,000 station-pollutant series, aiming to provide a comprehensive benchmark for air quality tasks. We benchmark this dataset across 11 leading time series foundation models and classical baselines to assess performance on short-term air quality forecasting. Our results demonstrate that TSFMs are effective zero-shot forecasters and consistently outperform classical baselines, with our top-performing model employing a cross-modal architecture that leverages a vision foundation model for time series forecasting. AQA is publicly released at AirQualityArena.github.io

---

### CruiseBench: A Real-Flight-Aligned N-CMAPSS Benchmark for Engine RUL Prediction

Authors: Pu Cheng, Qiang Miao
Published: 2026-07-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2607.19380v1

Remaining useful life (RUL) prediction estimates how long an engine can continue safe operation and is central to maintenance planning. N-CMAPSS extends C-MAPSS by simulating run-to-failure aero-engine trajectories using recorded real-flight profiles and retaining complete within-flight time series rather than cycle-level snapshots. However, this added realism reduces evaluation control because full-flight records increase data volume and entangle degradation cues with operating-regime variation, complicating preprocessing choices and direct comparisons of RUL modeling performance. To mitigate this issue, this paper proposes CruiseBench, a cruise-stage RUL benchmark derived from N-CMAPSS. It introduces CPM-N-CMAPSS (Cruising-Period Mask for N-CMAPSS), a mask artifact that stores cycle-local cruising intervals identified by the common-altitude method for the nine accessible subdatasets. CruiseBench applies a fixed protocol to the masked rows, using scenario descriptors and measured sensors as inputs while excluding virtual sensors, health parameters, and auxiliary metadata from the feature tensor, preserving native-resolution windows, and applying dataset-wise RUL caps. Experiments with LSTM, GRU, TCN, and TSMixer provide baseline results for this setting. Under CruiseBench-eta5-W256-S10, TSMixer obtains the lowest average RMSE, $3.4\pm1.71$, and Saxena score, $(2.50\pm2.99)\times 10^{4}$. Ablation studies show that flight-stage selection, temporal downscaling method, and RUL-cap threshold affect reported results. With its fixed cruise-stage protocol, CruiseBench provides a reproducible sub-benchmark for controlled RUL model comparison and CPM-N-CMAPSS provides a stage-specific data foundation for future transfer-learning and domain-adaptation studies.

---

### Bayesian Wind Tunnels for Model Selection

Authors: Siddhartha R Dalal, Vishal Misra, Abhay Parekh
Published: 2026-07-01
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2607.19379v1

Prior work has shown that transformers can perform exact Bayesian filtering within a fixed hypothesis class. Can they also perform Bayesian model selection -- identifying the correct hypothesis class from data? We introduce model-selection Bayesian wind tunnels: controlled environments where ground-truth posteriors over hypothesis classes are available in closed form. Using fixed-point-free involutions -- whose defining property f(f(x))=x is purely relational -- a 2.8M-parameter transformer achieves 0.01-bit entropy agreement with the Bayesian optimum (3 seeds), with both integer tokens and opaque symbols whose meanings change every episode. This extends to non-nested comparisons: involutions vs. 3-cycles (where neither class is a subset of the other) achieve class-posterior MAE under 0.001, demonstrating genuine model selection beyond simplicity/subset bias. We then identify a sharp perceptual access condition: when the discriminative statistic requires arithmetic -- modular addition (rotations) or multiplication (f(x)=cx mod p) -- model selection succeeds with integer tokens but fails completely with opaque symbols, and this boundary persists under 112x scaling (2.8M to 316M parameters). A stationarity control confirms the operative factor: opaque tokens with a fixed relabeling succeed (0.009-bit MAE), showing that stable semantics, not integer identity, enable circuit compilation. Header subtask diagnostics localize the failure to the composition of header inversion with arithmetic rather than header parsing itself. Probing frontier LLMs on the same tasks shows qualitative Bayesian behavior but a large calibration gap (~55x), measured through lossy probes and therefore directional rather than exact.

---

### Native Multi-Dimensional Subquadratic Operators via Input Dependent Long Convolutions

Authors: David R. Wessels, Farhad Ramezanghorbani, Alireza Moradzadeh, David W. Romero, Olivia Viessmann, Maksim Zhdanov, John St. John, Ken Janik, David M Knigge, Yucheng Tang, Erik J Bekkers, Saee Gopal Paliwal
Published: 2026-07-01
Categories: cs.LG, cs.CV, stat.ML
arXiv: https://arxiv.org/abs/2607.19378v3

Subquadratic alternatives to attention require compromises when applied to multi-dimensional data: standard convolutions lack global receptive fields and input dependency, while recurrent models require rasterizing data such as images, volumes, and partial differential equation (PDE) into an ad-hoc $1\rm D$ scan order that violates their spatial structure. We introduce \textit{HyenaND}, a subquadratic, global, input-dependent operator that acts directly on the native geometry of multidimensional data through convolutions with implicitly parametrized global, input-dependent multi-dimensional convolutional kernels. Our CUDA implementation, \texttt{nSubQ}, fuses the FFT-convolution path to turn HyenaND's $\mathcal{O}(L \log L)$ scaling into wall-clock speedups. Across long-context genomics, computer vision, medical imaging, and PDE modeling, pure HyenaND stacks match the accuracy of strong attention baselines, while hybrid configurations that interleave HyenaND and attention layers outperform both pure attention and strong recurrence-based hybrids.

---

### Native Multi-Dimensional Subquadratic Operators via Input Dependent Long Convolutions

Authors: David R. Wessels, Farhad Ramezanghorbani, David W. Romero, Alireza Moradzadeh, Olivia Viessmann, Maksim Zhdanov, John St. John, Ken Janik, David M Knigge, Yucheng Tang, Erik J Bekkers, Saee Gopal Paliwal
Published: 2026-07-01
Categories: cs.LG, cs.CV, stat.ML
arXiv: https://arxiv.org/abs/2607.19378v2

Subquadratic alternatives to attention require compromises when applied to multi-dimensional data: standard convolutions lack global receptive fields and input dependency, while recurrent models require rasterizing data such as images, volumes, and partial differential equation (PDE) into an ad-hoc $1\rm D$ scan order that violates their spatial structure. We introduce \textit{HyenaND}, a subquadratic, global, input-dependent operator that acts directly on the native geometry of multidimensional data through convolutions with implicitly parametrized global, input-dependent multi-dimensional convolutional kernels. Our CUDA implementation, \texttt{nSubQ}, fuses the FFT-convolution path to turn HyenaND's $\mathcal{O}(L \log L)$ scaling into wall-clock speedups. Across long-context genomics, computer vision, medical imaging, and PDE modeling, pure HyenaND stacks match the accuracy of strong attention baselines, while hybrid configurations that interleave HyenaND and attention layers outperform both pure attention and strong recurrence-based hybrids.

---

### Reliability-Aware Hard--Soft Physics-Informed Neural Networks for Robust Learning of Challenging Partial Differential Equations

Authors: Duc Tien Nguyen, Hang Tran, Trinh Minh Tuan, Nguyen Duc Manh, Dinh Gia Ninh
Published: 2026-07-01
Categories: math.NA, cs.LG, physics.flu-dyn
arXiv: https://arxiv.org/abs/2607.19377v1

Physics-informed neural networks (PINNs) provide a mesh-free framework for solving partial differential equations, but their training is often affected by loss imbalance, optimization stiffness, and difficulty in capturing localized or multi-mode solution structures. Hard-soft PINNs (HSPINN) alleviate part of this difficulty by embedding Dirichlet or periodic constraints directly into the trial space, but the resulting fixed admissible representation can still be poorly conditioned for sharp or heterogeneous residual fields. This paper proposes a reliability-aware hard-soft PINN (RA-HSPINN) that preserves exact embedded constraints while introducing a bounded learnable reliability field to modulate the interior representation. The method combines this reliability-aware ansatz with inverse-EMA global loss balancing and lightweight regularization, while retaining the standard mean-square residual form. The reliability field is a numerical modulation variable, not a physical parameter or calibrated probability. RA-HSPINN is evaluated on nonlinear Burgers equations, periodic convection, a mixed-boundary Poisson problem, and a mixed first-order Poisson system. Compared with HSPINN, it reduces the relative error by $98.65%$ for sharp-gradient Burgers, $72.42%$ for Burgers data with noisy and incompatible initial conditions, $61.18%$ for smooth periodic convection, $60.02%$ for localized periodic convection, $29.36%$ for mixed-boundary Poisson, and $82.17%$ for a multi-mode mixed first-order Poisson system. The results show that reliability-aware modulation is most beneficial when hard-soft trial spaces are admissible but difficult to optimize, especially in localized, unreliable-data, and multi-mode PDE regimes.

---

### On the Limits of Support-Preserving Alignment and Bounded Filtering

Authors: Aryan Dutt, Rui Mao, Anupam Chattopadhyay
Published: 2026-07-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2607.18295v2

We study whether alignment schemes that reshape a base model's output distribution, combined with bounded safety filters, can drive the probability of harmful behavior to zero in modern large language models. Recent research suggests that harmful behaviors can persist under preference-based alignment and that external filtering can be computationally hard in the worst case, but it remains unclear whether practical alignment pipelines that largely preserve internal representations can eliminate harmful behavior entirely rather than merely suppressing its most visible forms. We formalize this setting using support-preserving alignment operators together with bounded filtering algorithms under black-box, white-box, and statistical-query access, and analyze their ability to approximate an ideal eliminator that removes all harmful mass. Building on this framework, we provide computational and information-theoretic arguments indicating that, under these constraints, bounded filtering may fail to eliminate all harmful outputs supported by the base model's distribution. To evaluate these limits empirically, we analyze a range of state-of-the-art open-weight and hosted LLMs accessed via OpenRouter under bounded black-box, white-box, and statistical-query filters on adversarial prompts drawn from curated cybersecurity scenarios and PKU-SafeRLHF. Across models, filter classes, and query budgets, the estimated harmful-output rate decreases with additional filtering compute but consistently plateaus above zero, suggesting a persistent empirical harm floor.

---

### AV-JEPA: Extending LeJEPA to Audio-Visual Self-Supervised Learning

Authors: Benjamin Robson, Santeri Mentu, Wenshuai Zhao, Arno Solin
Published: 2026-07-01
Categories: cs.MM, cs.AI, cs.LG, cs.SD
arXiv: https://arxiv.org/abs/2607.15295v1

We present AV-JEPA, an elegant multimodal extension of LeJEPA to audio-visual self-supervised learning. Using an early-fusion Vision Transformer and modality dropout as masking, the model is trained to align the embeddings of global and per-modality local views, while the SIGReg objective encourages a theoretically optimal distribution. This achieves cross-modal alignment in the latent space, resulting in a remarkably clean architecture with no decoder, EMA teacher, complex multi-term losses, or contrastive negatives. The proposed AV-JEPA backbone delivers competitive classification performance on VGGSound (57.1% top-1) and AudioSet (32.7 mAP) and supports zero-shot audio-video retrieval out of the box.

---

### Precomputing the Future-Offset Average in TriAttention

Authors: Amarnath Mukherjee
Published: 2026-07-01
Categories: cs.DS, cs.LG
arXiv: https://arxiv.org/abs/2607.13051v1

TriAttention is a recent method for shrinking the KV cache of long-reasoning LLMs: it scores each cached key by how much attention it is likely to receive and evicts the lowest-scoring ones. Because a key does not know how far away its future queries will sit, the score is averaged over a ladder of 17 possible future distances. We point out that this average is free: the future distance enters the score only through the position-dependent rotation, so the whole 17-fold average collapses--exactly, by a one-line algebraic identity--into a single per-band weight that is computed once, offline. Scoring a key then costs one evaluation instead of seventeen, with no change to which keys get pruned. The saving is modest and lives entirely in TriAttention's pruning-score computation, not in the attention kernel; we present it as a small, exact complement to their method, and we confirm the identity numerically.

---

### In-Context Reinforcement Learning under Non-Stationarity: A Survey

Authors: A Run, Ziluo Ding
Published: 2026-07-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2607.11906v1

The development of decision-pretrained transformers, algorithm distillation, long-context meta-RL, and retrieval-augmented agents has renewed interest in in-context reinforcement learning (ICRL): the ability of a pretrained or fine-tuned decision model to infer latent task rules and improve future behavior from interaction context, without test-time parameter updates. This line of work asks when trial-and-error evidence, rewards, transitions, demonstrations, feedback, or retrieved experience can make learning-like computation happen inside the context window. However, existing surveys of ICRL mainly organize the field around pretraining objectives, architectures, context formats, evaluation protocols, and theoretical mechanisms, while the non-stationary setting remains comparatively underexamined. In changing environments, accumulated context is not merely more evidence about a fixed task: the reward specification, transition kernel, observation channel, action interface, constraint model, or demonstration and memory distribution can fall out of alignment with the current regime. Previously useful context can therefore become stale, misleading, or useful again when an old regime returns. We survey non-stationary ICRL as the problem of adapting through context while deployed policy parameters remain fixed: the policy must infer both the current decision rule and which parts of its accumulated evidence still support that rule. We define non-stationary ICRL, relate it to meta-RL, decision sequence modeling, retrieval-augmented RL, value- and model-aware ICRL, and reward-feedback agents, and organize the literature along three questions: what changes, how the change unfolds, and how observable the change is to the agent.

---

### CHARLIE: An On-Premise Multi-Agent Retrieval-Augmented Generation System for Evidential Reasoning in Forensic Science

Authors: Leandro D. Carneiro, Andre L. S. Meirelles, Juliano de A. Gomes, Rafael C. A. Cabral
Published: 2026-07-01
Categories: cs.DL, cs.AI
arXiv: https://arxiv.org/abs/2607.05428v1

We present Charlie, an on-premise multi-agent Retrieval-Augmented Generation (RAG) system for structured evidential processing in digital forensic environments. Contemporary forensic workflows must handle large volumes of heterogeneous and unstructured documents under strict requirements of traceability, confidentiality, and legal compliance. Charlie addresses this challenge through a controlled agent architecture that combines local retrieval, task decomposition, structured memory, and verification mechanisms. Unlike cloud-based systems, it operates entirely within institutional infrastructure, preserving data sovereignty and evidential integrity. We describe the systems architecture, including its transition from classical RAG to agent-based orchestration, and demonstrate its application in real-world forensic scenarios. Case studies show that Charlie enables scalable multi-document data extraction and supports longitudinal forensic intelligence generation while maintaining traceability and auditability. Our results indicate that agent-orchestrated, on-premise RAG architectures can effectively support evidential workflows without compromising legal and institutional constraints. Charlie provides a practical and reproducible blueprint for deploying AI systems in high-stakes forensic environments. This manuscript is an archival version of a paper presented at the RELAF 2026 Workshop.

---

### TAG: A Lightweight Framework for Test-Driven Agentic Artifact Generation

Authors: Yaniv Melamed, Yoni Zukerman, Michal Shechter, Miri Weissler, Ashwin Patil, Hani Neuvirth-Telem
Published: 2026-07-01
Categories: cs.CR, cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2607.02615v2

Generating structured artifacts with Large Language Models - e.g.\ database queries, threat framework mappings, entity schemas - is relatively straightforward; however, making them reliable enough for production deployments presents challenges. We present TAG, a lightweight framework based on a core principle: \textit{LLMs generate, we validate}. This reframing shifts responsibility from generation quality to validation rigor. The framework rests on three key attributes: First, \textbf{test driven generation}: when tests fail, the LLM receives indicative error messages that expose why the output failed, enabling the LLM to understand its mistakes and refine subsequent attempts. Second, \textbf{deterministic and LLM-based tests}: deterministic tests catch heuristics that can be programmatically verified (schema, syntax, cross-reference), while LLM-based tests evaluate nuanced semantic and delicate features that resist programmatic inspection (intent alignment, logical consistency, domain correctness). Third, \textbf{expert-distilled judges}: LLM-based tests are calibrated to distill and replicate human expert decision distribution, transforming manual human quality gates into scalable, reusable evaluation proxies that reflect professional-grade validation standards. We demonstrate the framework on three artifact types in the security domain - KQL query generation, MITRE ATT\&CK mapping, and entity mapping - deployed in production at Microsoft Sentinel. We believe this framework can be applied beyond security to other artifact generation tasks, providing a path to reliable, high-quality outputs without sacrificing the efficiency gains of LLM generation.

---

### Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers

Authors: Aravind Pradeep, Samira Nazari, Mahdi Taheri, Christian Herglotz
Published: 2026-07-01
Categories: cs.CV, cs.AI, cs.AR, cs.DC, cs.LG
arXiv: https://arxiv.org/abs/2607.02612v1

Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative. Recent adaptive inference methods reduce this cost by selectively compressing tokens or terminating inference early, but combining these mechanisms often causes unstable intermediate representations and accuracy degradation. We introduce Fusion, a unified adaptive inference framework that coordinates token merging, early exiting, and token pruning through a simple staged design: tokens are merged first, confidence is evaluated next, and pruning is applied only to samples that continue inference. This ordering allows the three mechanisms to operate cooperatively rather than competitively. Fusion further includes lightweight routing modules that adapt compression strength to each input and support inference-time adjustment of the accuracy--latency trade-off without retraining. On ImageNet-1k with DeiT-S, Fusion matches or surpasses state-of-the-art adaptive ViT methods at comparable compute budgets while reducing calibration error by up to $4\times$ and inference energy by $48\%$. Experiments across ImageNet-100, CIFAR-100, and ImageNette with multiple ViT backbones demonstrate consistent transferability without dataset-specific tuning.

---

### Privacy-Preserving Industrial Ergonomics: mmWave-Based Automated REBA Scoring and Pose Estimation

Authors: Xuhan Zhang, Zhuangzhuang Dai, Luis J. Mans, Victor Chang
Published: 2026-07-01
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2607.02611v1

Work-related Musculoskeletal Disorders (WMSDs) require continuous ergonomic assessments. While Rapid Entire Body Assessment (REBA) is a gold-standard observation tool, manual monitoring is labor-intensive, and vision-based automation leads to privacy concerns. This paper proposes a novel end-to-end multi-task learning framework for privacy-preserving ergonomic assessment using millimetre-wave (mmWave) radar. A spatio-temporal backbone reconstructs 3D human skeletons, which serves as the biomechanical foundation for a subsequent regression head to generate REBA risk scores. To overcome the sparsity of radar point clouds, we utilise a multi-objective loss function incorporating biomechanical limits and temporal smoothness constraints. Furthermore, we implement an oversampling strategy to address the imbalance of high-risk postures in existing datasets. Experimental results on MMFi dataset demonstrate that our framework achieves a Categorical Accuracy of 77.78% and real-time performance with an inference latency of 5.70 ms. Our method reaches a High-risk REBA MAE of 0.93, which significantly outperforms both direct regression and two-stage pipelines in high-risk scenarios, providing a robust solution for non-invasive industrial ergonomic assessment.

---