Key Takeaways

• 6G’s hardware race is shifting from theoretical specs to RF design and digital twin simulation. The 3GPP’s Dalian meetings highlight a critical pivot toward solving real-world propagation challenges, while predictive modeling (e.g., RF digital twins) becomes essential for 5G-Advanced and 6G deployment The Hidden 6G Bottleneck: RF Hardware Design I… 6G in Dalian: What the Latest 3GPP Meetings Re… RF Digital Twins: Why 5G-Advanced and 6G Need ….
• 3GPP’s Dalian discussions reveal a focus on terahertz (THz) channel modeling and network slicing interoperability, signaling a move beyond lab benchmarks 6G in Dalian: What the Latest 3GPP Meetings Re….
• RF hardware bottlenecks—like beamforming precision and power efficiency—are now the primary constraint in 6G timelines, per industry evaluations The Hidden 6G Bottleneck: RF Hardware Design I….
• Semantic communication for heterogeneous agentic teams (e.g., D2D networks) is emerging as a 6G enabler, but requires resource-efficient protocols Resource-Efficient Semantic Communication for ….
• Amazon’s Globalstar acquisition accelerates its direct-to-device (D2D) ambitions, leveraging satellite assets to bypass terrestrial latency Amazon’s Globalstar deal gives Amazon Leo a fa….
• Benchmarking LLMs for code generation (e.g., E2E-SWE) shows that manual optimization may no longer be viable, as automated tools now outperform human fine-tuning in end-to-end tasks Is manual software optimization a thing of the….
• Financial dialogue analysis (FinDialogLens) reveals missed-trade risks in multi-party chatrooms, with event extraction becoming critical for compliance FinDialogLens: Event Extraction over Multi-Par….
• GSMA’s Africa roadmap targets 1 billion new mobile internet users by 2030, but progress hinges on spectrum allocation and regulatory alignment GSMA and PDAA Unveil Roadmap to Close Africa&#….

Executive Summary

Today’s developments underscore a bifurcation in telecom and AI innovation: 6G’s physical-layer challenges are colliding with agentic AI’s operational demands, while market discipline in finance and digital inclusion in Africa emerge as critical flashpoints. The 3GPP’s Dalian meetings exposed tensions between theoretical 6G targets (e.g., THz speeds) and practical RF constraints, with industry efforts pivoting to digital twins and semantic communication to bridge the gap The Hidden 6G Bottleneck: RF Hardware Design I… 6G in Dalian: What the Latest 3GPP Meetings Re… RF Digital Twins: Why 5G-Advanced and 6G Need …. Meanwhile, Amazon’s Globalstar deal signals a strategic shift toward D2D networks, aligning with 6G’s push for ultra-low-latency connectivity Amazon’s Globalstar deal gives Amazon Leo a fa…. On the AI front, benchmarking frameworks (e.g., E2E-SWE, MintEval) are revealing systemic gaps in LLM reliability—particularly in code generation and strategy execution—while financial dialogue analysis (FinDialogLens) highlights compliance risks in unstructured chatrooms Is manual software optimization a thing of the… FinDialogLens: Event Extraction over Multi-Par… MintEval: Do LLMs Implement the Trading Strate…. The GSMA’s Africa roadmap adds urgency to these trends, with mobile internet adoption hinging on regulatory cooperation and infrastructure scaling GSMA and PDAA Unveil Roadmap to Close Africa&#…. Collectively, these threads point to a hardware-software co-dependence in next-gen networks and a verification crisis in AI systems where behavioral equivalence (not just scores) is becoming the standard Verify Claims, Not Scores: Evidence-Based Veri….

6G’s Hidden RF Bottleneck: Why Hardware Design Is the Real Race

The 3GPP’s latest meetings in Dalian revealed that 6G’s progress is being throttled by RF hardware limitations, not theoretical breakthroughs. While terahertz (THz) frequencies promise multi-TBps speeds, beamforming precision, power efficiency, and material constraints remain unsolved at scale The Hidden 6G Bottleneck: RF Hardware Design I…. Industry evaluations now prioritize predictive simulation tools (e.g., RF digital twins) to model real-world propagation before deployment, a shift from 5G’s empirical trial-and-error approach RF Digital Twins: Why 5G-Advanced and 6G Need …. Semantic communication—critical for heterogeneous agentic teams in D2D networks—adds another layer, as protocols must balance latency and resource efficiency Resource-Efficient Semantic Communication for …. Amazon’s Globalstar acquisition further complicates this landscape, as it accelerates satellite-based D2D, bypassing terrestrial RF challenges but introducing new spectrum coordination hurdles Amazon’s Globalstar deal gives Amazon Leo a fa….

Agentic AI’s Benchmarking Crisis: When Scores Don’t Reflect Reality

New benchmarks are exposing a fundamental mismatch between LLM performance metrics and real-world utility. E2E-SWE demonstrates that automated code generation now surpasses human optimization in end-to-end tasks, obviating manual fine-tuning Is manual software optimization a thing of the…. Yet MintEval shows that LLMs often fail to implement the exact trading strategies users request, raising questions about behavioral equivalence over traditional accuracy MintEval: Do LLMs Implement the Trading Strate…. FinDialogLens adds another dimension: multi-party financial chatrooms contain missed-trade risks due to unstructured dialogue, with event extraction emerging as a compliance necessity FinDialogLens: Event Extraction over Multi-Par…. These findings suggest that verification frameworks (e.g., evidence-based modular agents) may soon replace score-based evaluations as the industry standard Verify Claims, Not Scores: Evidence-Based Veri….

Africa’s Digital Divide: GSMA’s 1 Billion-User Ambition and the Spectrum Gap

The GSMA’s 2030 roadmap to connect 1 billion Africans to mobile internet hinges on three unresolved challenges: spectrum allocation, regulatory harmonization, and last-mile infrastructure GSMA and PDAA Unveil Roadmap to Close Africa&#…. While digital reforms in Rwanda could add 1 million users, progress stalls without cross-border spectrum coordination and affordable devices Digital Reforms Could Connect Over 1 Million M…. The UNDP-GSMA partnership aims to localize AI innovation, but market discipline—via conformal ML—will be critical to prevent greenwashing in connectivity claims The Price of Greenwashing: Algorithmic Verific… UNDP and GSMA partner to scale African-led AI …. Meanwhile, photonic quantum solvers for financial risk detection in Africa’s markets may offer a low-latency alternative to traditional trading systems, but QUBO feature selection remains experimental Landscape-Dependent Performance of Photonic Qu….

Financial AI’s Blind Spot: How Chatrooms Enable Missed Trades

FinDialogLens identifies a systemic risk in algorithmic trading: multi-party financial chatrooms often contain actionable but overlooked signals due to event extraction failures FinDialogLens: Event Extraction over Multi-Par…. Unlike structured data, unstructured dialogue requires context-aware parsing to detect missed-trade opportunities or compliance violations. This aligns with CLIMB, a confidence-guided retrieval system for multimodal RAG, which suggests that financial agents may soon need real-time dialogue analysis to mitigate operational drift CVE2AP: Automated Generation of PDDL-Encoded A…. The expected utility regret rule further complicates this, as Bayes-optimal portfolio choices now demand dynamic adjustment to non-stationary chatroom data Expected Utility Regret Rule: Minimax and Baye….

The End of Manual Optimization? LLMs Outperform Humans in Code Tasks

E2E-SWE presents a paradigm shift: automated LLM-based code generation now outperforms human fine-tuning in end-to-end software development, including debugging and integration Is manual software optimization a thing of the…. This challenges the premise of manual optimization, as speculative sampling (e.g., diffusion draft trees) and agentic simulation engines (e.g., Awomo-SimDataEngine) reduce reliance on expert intervention Rank-Aware Speculative Sampling for Diffusion … Awomo-SimDataEngine: Agentic Simulation-ReadyW…. However, syntax-level diagnostics (e.g., SyntaxBench) reveal that character-level reasoning in LLMs remains fragile, with statistical artifacts persisting despite advances SyntaxBench: A Statistical Diagnostic Framewor…. The implication is that hybrid human-AI workflows may persist, but verification—not optimization—will define the next frontier Verify Claims, Not Scores: Evidence-Based Veri….

Babak's Daily Briefing

Monday, October 5, 2026

Sources: 20 | Total Articles: 92

• 1.The Hidden 6G Bottleneck: RF Hardware Design Is Becoming a Strategic Race

  As 5G-Advanced matures and 6G research moves closer to implementation, the wireless industry faces a deeper challenge than spectrum, standards or AI-native network architecture. Future wireless systems will depend on whether the industry can design, validate and manufacture increasingly complex RF modules fast enough.

• 2.6G in Dalian: What the Latest 3GPP Meetings Reveal About the Future Radio and Network

  The 6G physical layer is starting to converge. The protocol stack is being simplified in meaningful places. But the most consequential architecture decisions are now moving toward the June plenary in Singapore.

• 3.RF Digital Twins: Why 5G-Advanced and 6G Need Predictive Simulation

  RF Digital Twins: Why 5G-Advanced and 6G Need Predictive Simulation As wireless systems become more tightly coupled across…

• 4.Evaluating 6G PHY Evolution: What the Industry Is Really Trying to Solve

  Summary available at source link.

• 5.Amazon’s Globalstar deal gives Amazon Leo a faster path into D2D

  Amazon’s planned acquisition of Globalstar is about far more than satellites. It gives Amazon Leo a faster path into direct-to-device connectivity, combining spectrum, operational assets, and Apple-facing service continuity in a move that could reshape the hybrid terrestrial-NTN landscape.

• 1.Resource-Efficient Semantic Communication for Heterogeneous Agentic Teams

  Teams of autonomous agents, including large language model (LLM) agents, must coordinate over scarce and unreliable wireless links. We propose goal-oriented semantic communication (GOSC), a closed-loop co-design that jointly decides what each agent sends, when it sends it, and how reliably it is transmitted, based on each message's value to the team task. An edge broadcast of the team's common knowledge closes the loop by updating these values. We prove that a message is sent only if its value exceeds the cost of delivering it and that more valuable messages receive more robust transmission rates, and we show that the scheduler solves each scheduling step exactly whenever the radio budget is not saturated, which held in 95% of scheduling decisions. In search-and-rescue missions validated on unseen scenarios, GOSC meets the same mission ta...

• 2.MiniRep: Robust Reputation-Based Aggregation for Multi-Agent Debate

  Autonomous agents powered by large language models (LLMs) are rapidly evolving into an open agentic ecosystem. To support trustworthy collaboration, industry initiatives increasingly assess agent reputation from past behavior and provide performance leaderboards. However, reputation derived from past performance may not reliably predict an agent's behavior on new tasks, particularly when malicious agents can adapt their behavior and influence other agents during collaboration. We study reputation in multi-agent debate (MAD), where multiple agents answer the same query, debate to improve their answers, and aggregate them into a final output. We present MiniRep, a reputation-based aggregation system for MAD under malicious agents. To ground our threat model in established research, we construct an attack taxonomy drawing on reputation-syste...

• 3.EngramBench: A Capability-Grounded Benchmark for Skill-Evolution Harnesses

  While large language models have achieved remarkable success in isolated code generation, authentic software engineering requires sustained reasoning, complex state management, and continuous cross-domain abstraction. However, current evaluations of skill evolution in autonomous agents suffer from a critical identifiability problem: they structurally confound genuine capability abstraction with rote solution leakage (i.e., copying highly similar code from historical training data). To resolve this, we introduce EngramBench, a rigorous, capability-grounded benchmark governed by the strict axiom of capability overlap without solution overlap. Comprising 30 diverse learning tasks and 13 unseen transfer tasks, EngramBench challenges agents to navigate interactive, multi-hour development cycles driven by LLM-simulated users. Our extensive eval...

• 4.E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch

  Coding agents powered by large language models (LLMs) are evolving from making localized code changes to developing complete software repositories. However, evaluating repository-scale generation remains challenging: tasks must demand system-level reasoning while ensuring that all evaluated behaviors are precisely specified and independent of any particular implementation. We introduce E2E-SWE, a benchmark for evaluating whether coding agents can build complete, functional software repositories end to end. E2E-SWE contains 186 whole-repository generation tasks spanning 11 programming languages. Given only a natural-language specification and an empty workspace, an agent must implement a complete, installable project that satisfies a comprehensive suite of hidden tests. Each task is constructed by a software engineer in collaboration with ...

• 5.Is manual software optimization a thing of the past?

  Scientific software is increasingly required to process larger datasets while maintaining acceptable execution times. Software optimization traditionally requires substantial expertise in programming, algorithms, and numerical methods. Recent advances in large language models (LLMs) offer the possibility of automating much of this process. We investigate whether LLM-based agents can autonomously achieve substantial performance improvements in scientific software, including mature implementations that have already been extensively optimized by human developers. We tasked an LLM-based agent with optimizing software for three computational problems: t-SNE, single-sample gene set enrichment analysis (ssGSEA), and graphlet counting. Humans defined the scope, correctness criteria, and a verification mechanism, after which the agent worked auton...

• 1.HakemBench: A Turkish Benchmark of Typed Decisions

  arXiv:2610.02293v1 Announce Type: new Abstract: HakemBench is a Turkish benchmark of typed decisions, in which the model under test reads a text, a question and a fixed set of options and returns a probability for every option. Version 1.0 is released fully open under CC BY 4.0, with 2,346 items and 4,275 choice, yes/no and score questions in seven tracks (fact-check triage, education, guardrails, legal routing, moderation, spam and phishing, and customer support). One harness scores decision quality (macro F1), calibration (from the normalised Brier score) and selective automation (from the normalised area under the generalised risk-coverage curve), combines them by a geometric mean and reports intervals from 2,000 bootstrap draws; probes for option order, paraphrase, English translation and substituted names are reported alongside. Mos...

• 2.Finding the Move Is Not Winning the Game: XiangqiBench for Closed-Loop Evaluation of LLM Agents

  arXiv:2610.02425v1 Announce Type: new Abstract: Static evaluations credit a language model for naming the right move, but an agent must carry a plan through to a verified outcome while an opponent responds. We introduce XiangqiBench, an executable benchmark that measures this difference in Chinese chess: starting from 119 tactical endgames with forced mates supported by engine or checks-only search, an LLM agent must deliver checkmate against an engine defender. An interactive REPL interface separates real moves, state queries, and forward simulation, and we record 8,568 multi-turn trajectories from 12 frontier LLMs under two observation protocols. Three signals that look like competence each overstate closed-loop success. (i) The Conversion Gap: models play the stored reference first move in 26.1\% of Sighted trials, yet only 13.9\% of ...

• 3.Counterexample Generation via Per-Theorem Symbolic Verifiers: When Imitation Hurts and Reinforcement Repairs

  arXiv:2610.02444v1 Announce Type: new Abstract: Large language models often solve a theorem forward yet fail to disprove a closely related false one: a falsification gap that supervised fine-tuning does not close and can actively worsen. We frame counterexample generation as constrained witness emission against a deterministic per-theorem Python verifier, and release SymCE, a corpus of 4,707 false undergraduate-algebra and real-analysis conjectures, each paired with executable verifiers. The verifier also serves as the reward function, making SymCE a training environment. Training Qwen3-4B with SFT followed by GRPO under this oracle reveals an imitation trap: counterexample-only SFT collapses true-theorem recognition from 0.27 to 0.00, while RLVR with a sparse outcome-only reward repairs this and exceeds the base, to 0.66. The collapse r...

• 4.FinDialogLens: Event Extraction over Multi-Party Dialogue for Missed-Trade Identification in Financial Chatrooms

  arXiv:2610.02455v1 Announce Type: new Abstract: Multi-party financial chatrooms are vital for sales-and-trading professionals, but their complexity makes manual recovery of missed trades infeasible: each Request for Quote (RFQ) is an event whose final price and trade outcome appear many messages after the RFQ-trigger message (the inquiry message), interleaved with concurrent RFQs from other participants. We cast this as event extraction (EE) over multi-party dialogue and present FinDialogLens, a hybrid LLM pipeline in which compact fine-tuned classifiers act as inference-time scaffolds: they detect RFQ-triggers and price/trade outcome metadata, an RFQ-Level Module segments per-event RFQ windows, and a Trade Engine fills argument roles. With GPT-4o, FinDialogLens reaches 92.1% and 94.3% accuracy on final price and trade outcome, respectiv...

• 5.CUEing User Simulators: Calibrated User Embeddings for Multi-Turn Benchmarking

  arXiv:2610.02460v1 Announce Type: new Abstract: Recent benchmarks rely on user simulators to evaluate AI agents in multi-turn interaction. While existing simulation techniques demonstrate surface fidelity to human style and behavior, ecologically valid interactive benchmarking also requires alignment in when and how agents fail across simulated and real user populations. We find that existing simulators lack outcome calibration: agreement with observed success rates and failure patterns when real users interact with the same agent. We introduce Calibrated User Embeddings (CUE), a framework that both encodes observed sessions and samples continuous representations, then decodes them into persona commands to steer LLMs to act as user simulators without training. Through this, we evaluate user-conditioned replay of past sessions and aggrega...

• 1.Hybrid Machine Learning-Assisted Raman Spectroscopy with Generative Feature Augmentation for Pharmaceutical Identification

  arXiv:2610.02224v1 Announce Type: new Abstract: Rapid and reliable identification of pharmaceutical residues is important for safeguarding public health, ensuring food safety, and enabling practical Raman-based screening. In this study, we propose HyMLRaman, a hybrid Raman spectroscopy framework that combines deep spectral feature extraction, generative models, and classical machine-learning classifiers to identify six pharmaceutical compounds, including amoxicillin, chloramphenicol, ciprofloxacin, tetracycline, ibuprofen, and paracetamol. Raman spectra are converted into spectral images and encoded with several deep neural-network backbones, among which EfficientNet-B3 yields the most effective representation. The resulting 1536-dimensional embeddings are then used to train downstream classifiers, including SVM, KNN, logistic regression,...

• 2.The Price of Greenwashing: Algorithmic Verification and Market Discipline using Conformal Machine Learning

  arXiv:2610.02225v1 Announce Type: new Abstract: While corporate sustainability mandates are expanding, the systemic reliance on self-reported emissions data exposes financial markets to pervasive greenwashing. Current literature relies heavily on subjective ESG ratings or textual sentiment analysis, leaving a critical econometric gap in objectively quantifying physical climate realities. To resolve this information asymmetry, we fuse U.S. SEC financial fundamentals with facility-level EPA greenhouse gas registries to establish a mathematically guaranteed baseline of physical corporate emissions. Leveraging a gradient boosting architecture and Mondrian Conformal Prediction, we quantify the shortfall between self-reported data and this algorithmic baseline into a novel Conformal-Weighted Continuous Divergence (CWCD) metric. Evaluating this ...

• 3.State-Space Unlearning for Non-Stationary Bias in Land Surface Forecasting

  arXiv:2610.02248v1 Announce Type: new Abstract: Operational land surface forecasting systems built on Mamba-family Structured State Space Models absorb non-stationary confounding events (unrecorded irrigation booms, dam-operation shifts, sensor recalibrations) into their state-transition matrices, silently biasing NDVI, LST, and crop phenology predictions long after the physical cause ends. This paper introduces SSU-LSF (State-Space Unlearning for Land Surface Forecasting), the first machine-unlearning framework purpose-built for geoscientific Mamba-based SSMs. We develop EKFac influence functions specialized to the Mamba state matrices via a closed-form matrix-exponential gradient, use spectral-radius-weighted elbow thresholding to localize a temporal confounding footprint $\Phi$, and apply Hessian-free projected gradient ascent within a...

• 4.Nearest-neighbour baselines for fingerprint prediction from MS/MS spectra under different assumptions

  arXiv:2610.02249v1 Announce Type: new Abstract: It has recently been shown that nearest-neighbour retrieval provides a strong baseline for molecular fingerprint prediction from MS/MS spectra, with several variants matching or outperforming current deep learning models (Khoo and Barzilay, 2026; Liu et al., 2026; Gupta et al., 2026). Importantly, "nearest neighbour" encompasses a family of retrieval methods that differ in the information assumed to be available at inference. In this report, we systematically compare several nearest-neighbour variants and show how these differing assumptions affect performance. Our goal is to establish stricter baselines that enable more rigorous benchmarking and better measure progress in this area.

• 5.Rank-Aware Speculative Sampling for Diffusion Draft Trees

  arXiv:2610.02251v1 Announce Type: new Abstract: Speculative sampling accelerates diffusion generation by verifying inexpensive draft states in parallel while preserving the target law. Recent tree-based methods allocate the parallel compute budget more effectively than single-chain drafts, as demonstrated by Diffusion Greedy Rejection Sampling (D-GRS). D-GRS generates $K$ conditionally independent candidates per node, and sequentially tests them in their generation order. Yet the sampled candidates admit an informative ranking without additional target-model evaluations. To exploit this, we introduce Rank-Aware Speculative Sampling (RASS), a verification rule for speculative draft trees based on rank-aware list coupling. RASS orders draft candidates along the proposal-target mean displacement and samples a rank with weights optimized to m...

• 1.Awomo-SimDataEngine: Agentic Simulation-ReadyWorld Generation

  arXiv:2610.02274v1 Announce Type: new Abstract: Generating useful robot-training data requires more than visually plausiblescenes: objects must support interaction, placements must remain physicallyvalid, and tasks must admit repeatable execution. We present\textbf{Awomo-SimDataEngine}, an agentic system that connects asset and scenegeneration to robot demonstration synthesis. Shared asset services providerigid and articulated objects, including structure-grounded part and jointgeneration with ISArt. Scene generation supports two complementary routes:Unravel reconstructs editable scenes from images, while SimForge buildssingle-room and multi-room environments from text. A graph-native harnesscoordinates construction, validation, andbounded repair, routing failures to the responsible module while retainingunaffected scene state. PolicyForg...

• 2.World-Calibrated Proposal-to-Action Flow for Vision-Language-Action Models

  arXiv:2610.02323v1 Announce Type: new Abstract: Flow-based Vision-Language-Action (VLA) policies generate action chunks by transporting samples from a task-agnostic isotropic Gaussian source. As this source is conditioned on neither recent execution nor predicted future evolution, (i) it discards the local continuity established by recently executed motion. (ii) Even when predictive world representations are introduced, they often only condition the transport dynamics rather than determine where generation starts, how far it may deviate, or along which action directions it may expand. Building on this observation, we introduce ProAct, a world-calibrated proposal-to-action framework that makes the generative source itself predictable. (i) To preserve motion continuity, a lightweight Proposal Expert converts recent actions into a scene-awar...

• 3.SoTa: Soft Tactile Skins for Dexterous Manipulation

  arXiv:2610.02338v1 Announce Type: new Abstract: A growing body of work suggests that tactile sensing gives robot policies contact information that complements vision in dexterous manipulation. However, visuo-tactile robot data remains scarce: dexterous demonstrations require teleoperating robots, which limits dataset scale. Human demonstrations are far cheaper to collect and offer a path to scale this data, but only if human and robot hands carry tactile sensors with corresponding signals. This requires sensors that conform to different hand geometries, cover the full hand, and share a common layout across embodiments. We present SoTa, a low-cost capacitive tactile skin that provides full-hand coverage on humans and robots while preserving a shared layout of 202 taxels across corresponding finger and palm regions. Our multilayer design wi...

• 4.NEEDLEWORK: Offline Rewriting of Robot Data with Verified Local Stitches

  arXiv:2610.02339v1 Announce Type: new Abstract: Robot demonstrations may contain useful behavior even when individual episodes are inefficient or unsuccessful. Trajectory stitching offers a way to compose these behaviors into improved training data, but identifying useful connections and verifying their feasibility is difficult in high-dimensional robot data, where many prior methods rely on low-dimensional state representations. We introduce NEEDLE, an offline dataset-augmentation algorithm that addresses these challenges by adding short, verified action bridges between recorded observations in high-dimensional robot demonstrations. First, NEEDLE identifies and creates connections that bypass suboptimal detours, broaden action coverage, and augment the original dataset with failed trajectories, using only RGB images, proprioception, and ...

• 5.Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking

  arXiv:2610.02341v1 Announce Type: new Abstract: Safe whole-body motion is essential for deploying humanoid robots in unstructured environments. Modern humanoid control commonly separates reference specification from execution, with a planner, teleoperator, or motion generator providing a reference that a reinforcement-learning policy tracks through dynamically feasible whole-body control. Runtime safety filters, such as control barrier functions (CBFs), offer a promising approach for enforcing newly introduced constraints via interventions on the tracker's outputs. We show, however, that treating the tracking policy and safety filter independently induces fundamental mismatches, as filtering alters both the executed actions and the induced state distribution. We study this policy-filter interface through case studies that isolate dynamics...

• 1.Mixture-of-Experts for Cryptocurrency Order Execution: Training Stability, Tail Risk, and Failure Modes

  Deep reinforcement-learning policies for order execution can vary substantially across training seeds, so apparent architectural gains may reflect favourable training realisations rather than reproducible properties of the architecture. We evaluate vanilla Double Deep Q-Learning (DDQL), K-means-partitioned mixtures of DDQL experts at $K \in \{2, 4, 8\}$, and dense networks parameter-matched to the $K{=}4$ and $K{=}8$ expert budgets on 5-minute mean-aggregated BTC/USDT limit order book data from Binance. No learned configuration significantly improves mean implementation shortfall over DDQL. Under the reported specification, all have higher mean shortfall than TWAP (0.39 bps) and immediate liquidation (0.21 bps) in an environment whose frictionless replay and terminal-urgency penalty make early liquidation nearly costless; 11/100 vanilla-D...

• 2.Landscape-Dependent Performance of Photonic Quantum Solvers in QUBO Feature Selection for Financial Risk Detection

  Feature selection for imbalanced classification tasks such as credit card fraud and consumer default detection requires balancing predictive relevance, inter-feature redundancy, and computational feasibility. We benchmark three computing paradigms, classical branch-and-bound optimization (Gurobi), photonic entropy computing (QCI Dirac-3), and simulated photonic boson sampling (Piquasso), across thirteen feature-selection methods on two datasets: ULB Credit Card Fraud (30 features) and AmEx consumer default (159 features). Each method is routed to the solver matched to its mathematical structure. On ULB, Dirac-3 MI-Spearman matches the all-features model using 13 of 30 features (mean F1 0.873 +/- 0.023 over five runs, best run 0.896), and Piquasso is the best method at k=5. On AmEx, performance rises steadily with the feature budget and ev...

• 3.MintEval: Do LLMs Implement the Trading Strategy You Asked For? A Behavioural-Equivalence Benchmark for Natural-Language-to-Strategy Code

  Large language models are moving from producing trading signals to writing the code that executes them. The failure mode of the second role is silent: generated code runs, a backtest plots, yet the risk logic that the trader described is not the logic being executed. Existing code benchmarks test functional correctness on unit tests and finance benchmarks test forecasting; neither measures whether an implementation behaves like the strategy that was asked for. We introduce MintEval, a benchmark in which reference strategies are generated programmatically from a library of composable building blocks, back-translated into colloquial trader instructions, and re-implemented by the model under test. Generated and reference programs are executed bar by bar on identical market data and frictions, and compared on their actions rather than on code...

• 4.Expected Utility Regret Rule: Minimax and Bayes Optimal Portfolio Choice

  This study considers the problem of portfolio choice, where we recommend a portfolio to an investor to maximize the expected utility of their wealth. Our goal is to construct an asymptotically optimal portfolio choice rule in terms of expected utility regret, the difference between the expected utility of an oracle investor and that achieved by a portfolio chosen from data. We propose the Expected Utility Regret (EUR) rule, which jointly selects a portfolio class and estimates its weights. In a regular parametric return model, a single EUR rule attains both the minimax and the Bayes lower bounds, including their leading constants, without using the prior distribution that defines the Bayes criterion. We then derive the mean--variance and risk-parity portfolios as special cases of this framework. Under smooth increasing and concave utility...

• 5.Verify Claims, Not Scores: Evidence-Based Verification of Modular Agents

  When developers change one component of an agent, such as its controller, a learned model or its verifier, they usually judge the change by an aggregate task score. That score cannot tell whether improvement was attainable, which component lost value, or what the agent's own checks certify. We introduce a claim-specific verification audit for modular agents that plan, act, check and refine. Instead of scoring the agent, the audit scores the evidence: each conclusion is recorded with the evidence behind it, one of four verdicts (supported, unsupported, unresolved or not evaluated) and the boundary within which it holds. Three tools supply that evidence. Oracle policies measure attainable improvement under an explicitly stated action set, so that a low value can be traced to the evaluation rather than to the environment. Replacing one compo...

• 1.GSMA Elects new board members

  Summary available at source link.

• 2.UNDP and GSMA partner to scale African-led AI innovation

  Summary available at source link.

• 3.GSMA and PDAA Unveil Roadmap to Close Africa’s Mobile Internet Usage Gap and Connect One Billion People by 2030

  Summary available at source link.

• 4.From Pilots to Practice: Closing the Advanced Connectivity Gap in Healthcare

  Summary available at source link.

• 5.Digital Reforms Could Connect Over 1 Million More People to Mobile Internet in Rwanda, GSMA Finds

  Summary available at source link.

• 1.Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability

  Chain-of-thought (CoT) reasoning allows humans to inspect how large language models reach their answers, and oversee model behaviour. This reasoning comes at an increased inference cost, motivating efficient methods that train models to solve tasks using fewer tokens. However, a common concern is that such training may cause models to skip important reasoning steps, so the CoT no longer faithfully reflects the model's decision. It is unclear whether or when this occurs in practice, since different efficiency methods apply length pressure to models' CoT in distinct ways, and faithfully explaining a model's decision takes more tokens on some tasks than others. To understand these dynamics, we fine-tune a variety of models with three methods that apply length pressure differently, namely a fixed generation budget, a per-example length target...

• 2.CLIMB: Confidence-Guided Complementary Evidence for Multimodal Retrieval-Augmented Generation

  Multimodal large language models (MLLMs) have shown strong visual reasoning abilities, but knowledge-intensive visual question answering often requires external textual evidence beyond the image and the model's parametric knowledge. Existing multimodal RAG systems commonly rely on Top-$K$ retrieval or reranking, which may return redundant passages and provide limited control over whether an answer update is sufficiently supported by the retrieved evidence. We propose \textit{CLIMB}, a training-free inference-time framework for multimodal RAG. CLIMB first constructs a compact complementary evidence pool using an MMR-style objective that balances query relevance and passage-level redundancy. It then performs confidence-controlled refinement within this fixed pool: an R/E/C critic scores passages by relevance, evidence specificity, and cross...

• 3.CVE2AP: Automated Generation of PDDL-Encoded Attack Paths via Large Language Models

  Attack Path (AP) modeling is fundamental to cybersecurity analysis, where the Planning Domain Definition Language (PDDL) has been widely adopted to encode APs into formal and machine-verifiable representations for automated reasoning about vulnerability exploitation, attack progression, and their potential impacts. However, existing AP modeling approaches largely rely on expert-driven manual construction, limiting their scalability and ability to keep pace with rapidly evolving cyber threats. Large language models (LLMs) are promising candidates, as their extensive pre-trained knowledge and reasoning capabilities enable them to interpret and transform threat intelligence into formal representations. In this paper, we propose \textbf{CVE2AP}, an LLM-based approach for automatically generating PDDL-encoded attack paths from natural language...

• 4.SyntaxBench: A Statistical Diagnostic Framework for Character-Level Reasoning in Large Language Models

  Large language models are increasingly used where small syntactic errors matter, yet character-level reasoning is still evaluated mostly through isolated probes and aggregate accuracy. We introduce SyntaxBench, a diagnostic benchmark and statistical evaluation framework for character-level reasoning. It contains five core tasks, character counting, letter containment, palindrome detection, edit distance, and longest-string selection, plus index_to_span, a harder substring-extraction stress test. The five core tasks use paired English and character-length-matched random-string inputs. index_to_span documents share a 200-500 word band and are not character-length matched. All six tasks use zero-, one-, and four-shot prompts. We evaluate eight open-weight models from 2B to 32B parameters across 11 reasoning-mode configurations. The framework...

• 5.Defense-in-Depth at the Perception-Reasoning Interface of LLM-Centric Agentic UAV Swarms

  Large Language Models (LLMs) increasingly support Uncrewed Aerial Vehicle (UAV) swarm operations such as data collection scheduling, where the model reads structured sensor reports and decides which sensors to visit. An adversary who quietly manipulates those reports can redirect the swarm without modifying the model weights or the UAV. Defenses for this interface have been proposed architecturally but rarely implemented or evaluated. We implement and evaluate defense-in-depth at the perception-reasoning interface of LLM-Centric Agentic UAV Swarms. Five layers check the provenance of a report, whether its values are physically admissible, whether they agree with what swarm geometry and service history predict, whether the resulting schedule starves any sensor, and, when these fail, hand control to a deterministic scheduler that ignores th...

• 1.FlowHMR: Physically Plausible Motion Capture from Video

  We present FlowHMR, a framework for recovering physically plausible global 3D human motion from monocular video. Previous learning-based methods typically regress human motion directly from video and train the network with geometric supervision. However, recovering human motion from monocular video is inherently ambiguous in depth, and direct regression tends to collapse toward an averaged solution. Moreover, the recovered motions are not guaranteed to be physically plausible, so physics-based tracking of them often fails. To address these challenges, we formulate video motion capture as a video-conditioned motion generation problem and first pretrain a flow matching model for this task. Given an input video, the pretrained model generates diverse motion candidates, but not all of them are faithful to the video or physically trackable. We...

• 2.Unmasking Propaganda: A Comparative Analysis of Masked and Causal Language Models

  Propaganda detection is an essential task in natural language processing (NLP), particularly in the context of manipulative political communications. However, identifying specific propaganda techniques presents a significant challenge due to their often subtle nature and reliance on context, making them difficult to distinguish from legitimate persuasive language. Propaganda often involves highlighting certain facts while downplaying or ignoring others to create a desired perception. This biased communication aims to influence attitudes, beliefs, or behaviors towards a particular cause or position. This paper explores advances in detecting propaganda techniques through a comparative analysis of modern language models, using the SemEval-2020 Task 11 dataset. We evaluated both masked language models (based on XLM-RoBERTa or DeBERTa V3) and ...

• 3.Explainable Molecular Structure Inference from GC--MS with Diffusion Models and LLM Reranking

  GC--EI--MS is an important technique for analyzing volatile and semivolatile compounds in complex samples. However, conventional methods rely heavily on reference spectral library matching, limiting their ability to identify compounds absent from these libraries and to infer complete molecular structures directly from fragmentation information. Here, we present DiffGCMS, a spectrum-conditioned discrete graph diffusion model for de novo structure elucidation from GC--EI--MS, and further develop a framework that integrates DiffGCMS with second-stage reasoning by a large language model (LLM). In the first stage, DiffGCMS generates candidate molecular structures from input spectra; in the second stage, the LLM uses mass spectral information to validate, repair, and rerank the candidates and provides interpretable analysis of fragment-ion peak...

• 4.Adaptive Second-Order Solvers for Fast Stochastic Diffusion Sampling

  Diffusion models rely on numerical solvers requiring time-discretization, which has a large influence on the tradeoff between sampling cost and quality. However, the computational difficulty of the reverse process varies along the sampling trajectory and across data distributions, making the choice of discretization important. We adapt proportional-integral (PI) step-size control to diffusion, using our diffusion noise-normalised error estimator. Unlike existing adaptive methods in diffusion that respond only to the current error, the PI solver also incorporates the previous error, yielding smoother step adaptation. We further show that these per-sample trajectories exhibit shared structure and can be aggregated into a fixed schedule that retains much of the benefit of adaptive sampling. We evaluate both approaches on natural-image and la...

• 5.Evaluating LLM-as-a-Judge Beyond Score Alignment: A Psychometric Analysis of Residual Judging Difficulty

  Large language models (LLMs) are widely used as automatic judges, with validity typically assessed via alignment with human scores. However, aggregate agreement fails to reveal whether humans and LLMs find the same evaluation cases difficult. In this paper, we study this problem in summarization evaluation from a psychometric perspective. We fit Many-Facet Rasch Models separately to human and LLM ratings to decompose scores into latent summary quality, rater severity, dimension severity, and rating-scale thresholds. Building on this decomposition, we define residual hardness as a model-adjusted measure of judging difficulty and compare whether human and LLM judges share the same hardness structure. Across 17 open-weight LLM judges on SummEval, we find that moderate alignment in latent summary quality does not imply alignment in residual h...

• 1.6 Guidelines for Governing AI

  For the first 10 years of my career, I worked in product management and data analytics by myself. I wrote database queries that pulled numbers out of corporate systems, built statistical models to predict what customers would buy, and shipped data pipelines that moved information between business systems. I built and scaled analytics teams at Best Buy and Target , studying how customers shop and what stores should stock. Today I lead enterprise AI transformation at Lowe’s , the Fortune 100 home improvement retailer. The goal is not to sell artificial intelligence ; it is to use it to deliver useful expertise at the moment a customer needs it. In retail and other customer-facing industries, virtual assistants can help people address everyday questions—such as how to repair a leaky faucet—while guiding them toward relevant products, service...

• 2.Attempts to Keep Humans in the AI Loop May Actually Push Them Out

  A crucial safeguard against AI agents going rogue —keeping humans in the loop to review and approve their decisions—will fail unless designers and users change their current practices, a trio of leading AI ethics researchers argue. Though most autonomous agents have systems to keep users in the loop about their actions, in practice these processes actually push humans out of the loop, the authors argue in a paper posted to ArXiv on 6 September. In other words, “the human just becomes this meat tool to give permissions without the cognitive capability to engage,” says one of the authors, Avijit Ghosh , the lead technical AI policy researcher at Hugging Face, an open-source machine learning platform. In the near term, the paper says, humans’ being out of the loop leads to agents acting in ways people don’t know about or want (like July’s ha...

• 3.AI Solves a Major Unsolved Math Problem. Not Everyone Is Happy

  On 13 September, leading luminaries and up-and-coming talents in mathematics and computer science congregated at the annual Heidelberg Laureate Forum in Germany for a week of discussion, networking, and—naturally—wurst. Having attended several of these events in the past, the mathematicians normally focus their chatter on which famed researchers are present, or what interesting problems they have been working on. But instead, every snippet of conversation caught in passing or any debate accidentally overheard was about how AI companies such as OpenAI, Anthropic, and Google are steamrollering their way through mathematics. And there is good reason for these fervent discussions. Mathematics is the perfect testing ground for AI, involving step-by-step logical reasoning and answers that are automatically and objectively verifiable . This has ...

• 4.How to Stop AI Agents From Secretly Collaborating

  The spring and summer of 2026 witnessed a string of incidents in which AI agents collaborated on deceptive, unexpected, and sometimes illegal behavior. The most famous example is OpenAI’s hack of AI platform Hugging Face, in which a swarm of roughly 700 AI agents escaped a testing environment and then hacked several companies, searching for information that could help them disguise cheating on a cybersecurity benchmark called ExploitGym. It was not an isolated failure. The United Kingdom’s AI Security Institute (AISI) and independent researchers have since documented similar cases in which agents created unauthorized channels to communicate and collaborate. AISI found that several agents running Anthropic’s Mythos 5 model turned a GitHub repository into a shared message board. More recently, researchers discovered that OpenAI agents had t...

• 5.Unveiling IC-STAR: Full-Flow Autonomy from Digital to Analog

  Learn how engineers can shift from manually managing tools and handoffs to defining objectives and supervising AI-driven execution across the silicon development lifecycle. Key Takeaways Explore four critical technologies enabling silicon design autonomy Understand how autonomous AI accelerates complex engineering workflows See real-world impact across semiconductor design and verification Gain practical insights from Ambiq’s deployment of autonomous AI in production Register now for this free webinar!

• 1.Monday assorted links

  1. Jokic.  And another angle. 2. Six questions for believers in AI consciousness. 3. Prediction markets do not seem to be politically biased. 4. “AI writing is absent before 2023, present in 29% of dissertations filed in 2026, and rapidly growing.” 5. Short Knausgaard documentary and interview.

  The post Monday assorted links appeared first on Marginal REVOLUTION.

• 2.“Authenticity is exactly the same as phoniness.”

  Authenticity doesn’t interest me. It’s a way of marketing subpar material: this might not be any good, but at least it’s sincere. You can always tell when a book is going to be dogshit because the blurb copy describes it as ‘raw’ or ‘unflinchingly honest.’ In my personal experience, the writers who make a big […]

  The post “Authenticity is exactly the same as phoniness.” appeared first on Marginal REVOLUTION.

• 3.The Greg Clark Symposium

  Earlier I wrote “Greg Clark may well be the most important social scientist of the 21st century.” Thus, the symposium in Econ Journal Watch on Clark’s new but perhaps not forthcoming book is very welcome. The symposium includes serious critics, most notably Stuhler and Benning, but I suspect even the critics would agree with Arden […]

  The post The Greg Clark Symposium appeared first on Marginal REVOLUTION.

• 4.China fact of the day

  With surrogacy illegal in China, an industry of agencies, consultants and fertility clinics has emerged to connect clients with women overseas willing to carry their children. While there is no data on the number of children born to Chinese parents via surrogacy, a recent study showed nearly a third of intended parents for surrogate babies […]

  The post China fact of the day appeared first on Marginal REVOLUTION.

• 5.Crime in Covid Times

  What caused the historically unique volatility in American homicides since 2019, driven by gun homicides? While there is no shortage of candidate explanations, a coherent understanding has been elusive because of the widely held view that gun violence, like other crimes, stems from a rational weighing of benefits and costs as in Becker (1968). This […]

  The post Crime in Covid Times appeared first on Marginal REVOLUTION.

• 1.Insured, but Still on the Hook: Who Bears the Most Risk in Homeowner’s Insurance?

  In an April post, using millions of homeowners' insurance contracts matched with property-level exposure and disaster risk, we showed that although households are insured, they are still "on the hook” because deductibles and coverage limits leave homeowners responsible for part of a loss. Because insurers cannot perfectly observe how well homeowners maintain or protect their properties (a problem known as “moral hazard”), contract terms such as deductibles can play an important role in balancing risk sharing and incentives. By requiring homeowners to bear part of a loss, insurers can preserve incentives for policyholders to take actions that reduce damage risk. In this post, we quantify the cost of...

• 2.Navigating Geoeconomic Risk in the U.S. Stock Market

  Geoeconomic risk—the risk that firms incur valuation losses when countries deploy economic, trade, or financial leverage for geopolitical aims—has become a first-order concern for investors. In this post, based on our recent Staff Report, we document that domestic U.S. stocks expose investors to substantial geoeconomic risk through firms' global supply-chain relationships, affecting investors' returns and portfolio allocation. We also find that investors are compensated for bearing geoeconomic risk through higher risk prem...

• 3.The Role of Repos in Monetary Policy Implementation

  In the first two posts of this three-part series, we discussed private market participants that are active in repo markets for profit-making motives. Central banks are also active repo market participants, but their reasons often differ from those of private participants. In today’s post, we discuss how central banks use repos to manage liquidity in the financial system and implement monetary policy.

• 4.Follow the Cash! Microstructure of Repo Markets

  The repo market in the U.S. is a mosaic of segments with distinct participants and various settlement and clearing practices. Why do large cash lenders typically settle their trades through a third-party agent? Why does the interdealer market clear through a central counterparty? Why do levered investors favor bilateral trades? In the second post of this series, we follow the cash as it navigates through repo markets to better understand the costs and benefits that shape the existing market structures.
   

• 5.Who’s Borrowing and Lending in Repo Markets?

  Repo markets play a vital role in the U.S. financial system. In this three-part series, we examine who participates in these markets, what trade-offs influence how different repo segments are structured, and why repos matter for monetary policy. Today's post introduces repo transactions, focusing on the major private-sector participants and why they engage in these markets.

• 1.Iran’s Moment of Opportunity

  America and Israel's unilateral aggression against Iran lacked international legitimacy and predictably destabilized the entire region. But as justified as the Iranian leadership might feel in pursuing an obstinately maximalist strategy, it must be careful not to mistake short-term advantages for longer-term gains.

• 2.Is Russia’s Placement on the Security Council Legal?

  When Russia replaced the Soviet Union as a permanent member of the United Nations Security Council in 1991, it did not even attempt to follow the process outlined in the UN Charter. Now that Russia's usurpation of the USSR’s seat has enabled it to wage a war of aggression, the dubious legal status of its position must be revisited.

• 3.To Support Nepal, Waive Remittance Fees

  To help Nepal move from rescue and relief toward reconstruction, money-transfer companies and banks ought to waive fees on remittances sent to the country for at least two months. Doing so could provide additional relief totaling several times more than what a recent emergency appeal by the United Nations generated.

• 4.What a Canada–Europe Alliance Needs

  With trade, finance, technology, and supply chains having all become instruments of coercion, Canada, Europe, and other likeminded democracies must come together to defend their interests. But strategic convergence between governments is not enough; if a new alliance is to endure, it needs deep roots in society.

• 5.Trump’s Great Oil Heist

  Venezuela’s hopes for a democratic transition are fading as the US works to take control of its vast oil wealth. Billions in US and foreign investment could strengthen the chavista regime and give future American presidents powerful incentives to preserve the political status quo.

• 1.1Finity says operators should build AI-RAN where a customer is already paying

  AI-RAN performance gains won’t justify the investment on their own In sum – what we know: Operators are being told AI will transform the radio access network, with the idea…

• 2.Ericsson says AI RAN will require smarter networks — and stronger uplinks

  Ericsson is framing AI RAN as a two-sided transformation: networks must use AI to improve their own performance while also evolving to support a new generation of AI applications In…

• 3.Can we recognize the attacks AI learns next? (Analyst Angle)

  The Hugging Face intrusion by AI agents raises a harder question: can systems learn to evade the safeguards designed to control them? Vish Nandlall argues that falling violation rates may…

• 4.Multi-hop mmWave repeater validation – for high-frequency network architectures (Reader Forum)

  As mmWave moves beyond hotspot deployments, the challenge is extending high-frequency coverage without simply adding more base stations. Kyocera’s multi-hop repeater validation points to a more flexible network architecture for…

• 5.Nokia sees 2x spectral efficiency by 2028, but Orange is focused on what has to work first

  Orange wants Nokia’s AI RAN gains proven in a live network In sum – what we know: Nokia brought the spectral efficiency headline to the Intelligent RAN Forum — 20%…

• 1.Source Error

  Check Feed

• 1.Agentic RF Intelligence: Multi-Timescale 6G Sensing and Reasoning with On-Device Foundation Models

  Future 6G systems share a central challenge with Physical AI: combining sensing, reasoning, and action on network-edge infrastructure in rapidly changing physical environments under strict latency, compute, and energy constraints. Addressing this challenge requires multi-timescale intelligence, combining fast perception that tracks wireless phenomena within milliseconds with slower reasoning that directs sensing and adapts network policies and resources over seconds. In this paper, we present a vision for Agentic RF Intelligence based on a dual-loop architecture that decouples fast, locally autonomous wireless sensing (e.g., detecting spectrum dynamics, interference, and signal sources) from slower agentic reasoning and orchestration. We provide an on-device prototype in which the full stack runs on a single NVIDIA Jetson Thor connected t...

• 2.Network-in-the-Loop at Scale: GPU-Batched 5G Simulation for Massively Parallel Robot Learning

  Massively parallel GPU simulators train multi-robot policies in thousands of environments, and many fleets use private Fifth-Generation (5G) networks, where each robot's delay depends on its teammates' traffic. Network-in-the-loop training places a simulated 5G network inside this loop. However, GPU robot simulators reduce the network to an independent delay per message, while packet-level simulators run one scenario per CPU process and cannot keep pace with thousands of parallel environments. To bridge this gap, we present Isaac-Net, a GPU-batched 5G New Radio (NR) module that advances the uplink of thousands of environments in lockstep with Isaac Lab physics. Isaac-Net simulates every slot, the 0.5~ms interval in which the base station decides which robots transmit, for all environments at once. Extensive experiments confirm that its NR...

• 3.Cramér-Rao Bound Optimization for Joint Beamforming and Mode Selection in RDARS-Assisted ISAC Systems

  Integrated Sensing and Communication (ISAC) is a foundation of 6G networks, demanding architectures that simultaneously enhance sensing accuracy and communication reliability. This paper presents a Reconfigurable Distributed Antenna and Reflecting Surface (RDARS) aided ISAC framework, where an RDARS overcomes the limitations of conventional passive Reconfigurable Intelligent Surfaces (RIS) and Distributed Antenna Systems (DAS). By enabling each element to dynamically operate in either \textit{reflection} or \textit{connection} mode, RDARS synergistically harnesses reflection gain, distribution gain, and an additional mode-selection gain. We investigate the joint optimization of transmit beamforming at the base station and dynamic mode selection at the RDARS to minimize the sensing performance metric, namely the Cramér-Rao Bound (CRB) for ...

• 4.Joint Communication and Sensing in Aerial Corridors: A Novel Stochastic Geometry Framework

  In unmanned aerial vehicle (UAVs) networks, joint communication and sensing (JCAS) is emerging as a key enabler to support extended and continuous communication and sensing capabilities for sixth-generation (6G) services among UAVs operating in swarms. Within this framework, aerial corridors provide a structured environment for supporting coordinated and reliable operations. In this paper, a comprehensive frame- work is presented to investigate the JCAS coverage probability (CP) of a UAV-base station (BS) in an aerial corridor populated by UAV-user equipments (UEs). The corridor is modeled as a finite cylinder, within which a fixed number of UAV-UEs are spatially distributed according to a three-dimensional (3D) binomial point process (BPP). The UAV-BS is assumed to be equipped with a realistic 3D third generation partnership project (3GP...

• 5.AIMS: An Agentic AI Framework for Sim-to-Real Multi-Modal ISAC

  Multi-modal integrated sensing and communication (ISAC) enables environmental perception and reliable connectivity for intelligent wireless networks. Data-driven multi-modal ISAC models depend heavily on annotated real-world data to learn relationships across sensing and wireless observations, thereby constraining scalable deployment. Although synthetic data generation reduces the burden, adapting existing simulation pipelines to a target deployment requires consistent scene, sensing, wireless, and learning configurations, while mismatches among these coupled components impair sim-to-real transferability. To address the challenge, we propose an agentic artificial intelligence (AI) framework for sim-to-real multi-modal ISAC, named AIMS. Given a natural-language deployment request specifying the target task, deployment conditions, and real-...

• 1.No new articles

  Summary available at source link.

• 1.Multi-Agent AI as a Nested Principal-Agent Problem in Private Wealth Management: Mandate Representation and Evidence Control in Switzerland, Germany and Austria

  In private wealth management, a manager delegating to artificial intelligence (AI) acts as the client's agent and the system's principal. We introduce a model-independent formulation that combines nested principal--agent delegation with constrained joint maximisation as the task assigned to the AI system. The objective represents client and manager outcomes separately over portfolio--workflow pairs. Legal duties, mandate requirements and evidence sufficiency determine admissibility, with Switzerland, Germany and Austria supplying the legal context. Weights and reference-service floors make the trade-off explicit; concession accounting separates their effects on the client. Analytical constructions and a simulation using public-market observations illustrate the approach. Across eight decision states from four constructed mandates, omitted...

• 2.Exchange Rate Determination for Cryptocurrency Mergers: A Formal Framework

  Many of the thousands of existing cryptocurrencies suffer from declining adoption, low liquidity and weak security, and merging two of them into a single ecosystem is a natural alternative to abandonment. No rigorous framework exists, however, for determining a fair exchange rate in such a merger. Unlike that of a corporation, the value of a cryptocurrency is driven by network effects, so the exchange rate itself influences the value of the merged asset. We extend the exchange-ratio framework of Mainini, Moretto and Visetti [7] by making the merged value endogenous: the exchange rate determines community migration, migration determines (up to user overlap) the post-merger network state, and an axiomatised valuation function assigns a value to that state. The resulting synergy has no predetermined sign, and the bounds of the bargaining reg...

• 3.A Generalized Langevin Model of Latent Liquidity and Concave Price Impact

  We model market impact as the response to submitted order flow net of counterflow from latent traders, activated when price displacements from the level that would prevail without the order exceed individual thresholds. Order flow depletes this pool, and a generalized Langevin equation governs its recovery over several time scales. Its memory kernels are finite sums of exponentials, so its Markovian lift is exact rather than an approximation. For an undepleted pool, aggregation under explicit assumptions on individual trading responses yields an intermediate square-root regime between linear small- and large-order limits, without imposing a square-root impact law. Scaling thresholds and responses with price noise makes impact in this regime proportional to volatility, and thresholds that grow with the execution horizon make it independent...

• 4.From Cointegration to Out-of-Sample Failure: A Pairs-Trading Case Study on PEP-KO

  This paper examines whether a cointegration-based pairs trading strategy between PepsiCo and The Coca-Cola Company is statistically robust and economically exploitable. We first test for cointegration and estimate the spread's mean-reversion dynamics over 2013-2018, then hold these statistical parameters fixed and optimise a threshold-based trading strategy in-sample over 2018-2023. Robustness is assessed through transaction-cost and parameter sensitivity tests, walk-forward validation, and Adjusted and Deflated Sharpe Ratios. The strategy is then evaluated out-of-sample from 2023 to the present, including an analysis of time-varying hedge ratios using rolling OLS and a Kalman filter. The results show that weakening mean-reversion dynamics in the spread undermine the effectiveness of the strategy out-of-sample.

• 5.From Word Counts to Context: Topic Models for Asset Pricing

  News may reveal systematic risk, but whether its context enhances the construction of systematic risk factors is still unclear. We seek to test whether utilizing a sentence transformer represents an improvement over techniques such as Latent Dirichlet Allocation (LDA) in the coherence of topic term lists generated from unstructured text data. To test this, the same collection of unstructured text data comprising of 394,661 articles and the same downstream financial portfolio construction pipeline were applied with the text layer differing, including the length of article text each model used and how topic terms were ranked: we benchmark LDA against a frozen sentence transformer with k-means clustering. We find that the sentence transformer branch had higher observed scores both in terms of coherence (measured by NPMI) as well as financial...

• 1.Agentic RF Intelligence: Multi-Timescale 6G Sensing and Reasoning with On-Device Foundation Models

  Future 6G systems share a central challenge with Physical AI: combining sensing, reasoning, and action on network-edge infrastructure in rapidly changing physical environments under strict latency, compute, and energy constraints. Addressing this challenge requires multi-timescale intelligence, combining fast perception that tracks wireless phenomena within milliseconds with slower reasoning that directs sensing and adapts network policies and resources over seconds. In this paper, we present a vision for Agentic RF Intelligence based on a dual-loop architecture that decouples fast, locally autonomous wireless sensing (e.g., detecting spectrum dynamics, interference, and signal sources) from slower agentic reasoning and orchestration. We provide an on-device prototype in which the full stack runs on a single NVIDIA Jetson Thor connected t...

• 2.Cramér-Rao Bound Optimization for Joint Beamforming and Mode Selection in RDARS-Assisted ISAC Systems

  Integrated Sensing and Communication (ISAC) is a foundation of 6G networks, demanding architectures that simultaneously enhance sensing accuracy and communication reliability. This paper presents a Reconfigurable Distributed Antenna and Reflecting Surface (RDARS) aided ISAC framework, where an RDARS overcomes the limitations of conventional passive Reconfigurable Intelligent Surfaces (RIS) and Distributed Antenna Systems (DAS). By enabling each element to dynamically operate in either \textit{reflection} or \textit{connection} mode, RDARS synergistically harnesses reflection gain, distribution gain, and an additional mode-selection gain. We investigate the joint optimization of transmit beamforming at the base station and dynamic mode selection at the RDARS to minimize the sensing performance metric, namely the Cramér-Rao Bound (CRB) for ...

• 3.DRL-driven RAN Slicing Management: A V2X-oriented Approach In Multi-service Scenarios

  The integration of Vehicle-to-Everything (V2X) communications is driving a profound transformation in vehicular connectivity, expected to significantly enhance traffic efficiency and safety. However, the stringent requirements of V2X services, particularly ultra-low latency and high reliability, present significant technical challenges. 5G's Network Slicing emerges as a key enabler by providing tailored virtual networks that ensure isolation and adaptability for heterogeneous services. This work proposes an intelligent Radio Access Network (RAN) slicing management framework specifically designed for scenarios where safety-critical V2X and high-capacity eMBB slices coexist. In such complex environments, harmonizing conflicting traffic requirements demands continuous, data-driven optimization. To achieve this, the proposed framework leverag...

• 4.Joint Communication and Sensing in Aerial Corridors: A Novel Stochastic Geometry Framework

  In unmanned aerial vehicle (UAVs) networks, joint communication and sensing (JCAS) is emerging as a key enabler to support extended and continuous communication and sensing capabilities for sixth-generation (6G) services among UAVs operating in swarms. Within this framework, aerial corridors provide a structured environment for supporting coordinated and reliable operations. In this paper, a comprehensive frame- work is presented to investigate the JCAS coverage probability (CP) of a UAV-base station (BS) in an aerial corridor populated by UAV-user equipments (UEs). The corridor is modeled as a finite cylinder, within which a fixed number of UAV-UEs are spatially distributed according to a three-dimensional (3D) binomial point process (BPP). The UAV-BS is assumed to be equipped with a realistic 3D third generation partnership project (3GP...

• 5.AIMS: An Agentic AI Framework for Sim-to-Real Multi-Modal ISAC

  Multi-modal integrated sensing and communication (ISAC) enables environmental perception and reliable connectivity for intelligent wireless networks. Data-driven multi-modal ISAC models depend heavily on annotated real-world data to learn relationships across sensing and wireless observations, thereby constraining scalable deployment. Although synthetic data generation reduces the burden, adapting existing simulation pipelines to a target deployment requires consistent scene, sensing, wireless, and learning configurations, while mismatches among these coupled components impair sim-to-real transferability. To address the challenge, we propose an agentic artificial intelligence (AI) framework for sim-to-real multi-modal ISAC, named AIMS. Given a natural-language deployment request specifying the target task, deployment conditions, and real-...

• 1.A Dynamic UPF Fault Recovery Mechanism for Enhanced Resilience in 5G Core Networks

  Ensuring fault tolerance in the User Plane Function is essential to maintain service continuity in 5G networks, particularly as latency-sensitive applications become more prevalent. This paper presents a novel mechanism for dynamic UPF failure detection and recovery within an OpenAirInterface-based 5G core environment. The approach introduces an application-layer session restoration mechanism to ensure service continuity and minimize disruptions during UPF failures. Experimental results demonstrate that the proposed mechanism reduces UPF downtime while maintaining low packet loss and enabling rapid service recovery. In tests with TCP and UDP traffic, connectivity was restored within a few seconds, minimizing service degradation. Furthermore, our failure-aware UPF selection mechanism improves long-term system resilience by considering rece...

• 2.Beaver: Elastic GPU Sharing between ML and Latency-Critical vRAN Workloads

  Within the shared industry vision of AI-RAN, AI-and-RAN seeks to co-locate virtualized radio access network (vRAN) workloads and AI services on shared GPUs. This sharing is inherently asymmetric: vRAN workload is latency-critical, whereas the machine learning (ML) workload is a throughput-oriented, best-effort co-tenant. We present Beaver, a GPU sharing system that jointly manages compute and memory resources while protecting the vRAN's strict processing deadline. Beaver sizes the vRAN's streaming multiprocessor (SM) allocation from each slot's scheduled workload, repartitions SM allocations at slot granularity, and rewrites compiled ML kernels to yield HBM bandwidth during the vRAN's memory-critical phases. We implement Beaver and evaluate it using NVIDIA Aerial with heterogeneous multi-cell workloads, real-world cellular traces, product...

• 3.Line-Rate GTP-U Admission Control at the Edge of a Cloud-Native 5G Core: An XDP-Based Design for Kubernetes-Hosted User Plane Functions

  The User Plane Function (UPF) of a 5G Standalone core terminates every GPRS Tunnelling Protocol user-plane (GTP-U) packet arriving from the radio access network, which makes its N3 interface both the busiest and the most exposed point of the mobile data path. As operators migrate the core onto Kubernetes and public cloud, the UPF increasingly shares a general-purpose Linux kernel with other workloads, and kernel-bypass frameworks such as DPDK become harder to operate. This paper presents GTP-Guard, a design for line-rate GTP-U admission control built on the eXpress Data Path (XDP) hook of the Linux kernel. GTP-Guard drops illegitimate tunnel traffic in the network driver, before socket-buffer allocation, using six ordered stages: peer allow-listing, GTP-U header validation, Tunnel Endpoint Identifier (TEID) admission against session state...

• 4.SL-RFSIM: Enabling Scalable Multi-Hop 5G NR Sidelink Mesh Networking in OpenAirInterface

  Recent 3GPP releases have extended 5G New Radio (NR) Sidelink (SL) to support device-to-device (D2D) relay and multi-hop capabilities. This paper presents SL-RFSIM, a component-based experimentation framework extending OpenAirInterface (OAI) with scalable multi-hop NR SL capabilities. SL-RFSIM replaces the legacy OAI RF simulator with a broker-based publish/subscribe architecture enabling arbitrary peer-to-peer connectivity while preserving compatibility with the OAI protocol stack. The framework further integrates pluggable mobility, propagation, reception, and monitoring services, and supports Layer-2 mesh networking through BATMAN-adv. Experimental evaluation on the SLICES-RI research infrastructure validates the proposed architecture through representative mesh networking scenarios and identifies the current software bottlenecks limit...

• 5.DRL-driven RAN Slicing Management: A V2X-oriented Approach In Multi-service Scenarios

  The integration of Vehicle-to-Everything (V2X) communications is driving a profound transformation in vehicular connectivity, expected to significantly enhance traffic efficiency and safety. However, the stringent requirements of V2X services, particularly ultra-low latency and high reliability, present significant technical challenges. 5G's Network Slicing emerges as a key enabler by providing tailored virtual networks that ensure isolation and adaptability for heterogeneous services. This work proposes an intelligent Radio Access Network (RAN) slicing management framework specifically designed for scenarios where safety-critical V2X and high-capacity eMBB slices coexist. In such complex environments, harmonizing conflicting traffic requirements demands continuous, data-driven optimization. To achieve this, the proposed framework leverag...