Babak's Daily Briefing

Monday, October 19, 2026

Sources: 17 | Total Articles: 63

More Developments

• 1.Embeddings Are Overpriced? A Frozen 16 MB Vocabulary for Small and Ternary Language Models

  arXiv:2610.22428v1 Announce Type: new Abstract: In a small language model the vocabulary is the single largest part. In Gemma 3 270M the embedding table holds 168M of 268M parameters, and because quantisation methods leave the embedding in 16-bit, it is also most of the file once the rest of the model is compressed. We ask what that table actually buys. We replace it with one frozen table of 512-bit codes, 16 MB, with no trainable parameters, derived once from a pretrained embedding by principal-component projection and sign binarisation, and we train the same Gemma-shaped body from scratch on the same 2.05B tokens with and without it. The learned table buys 0.20 nats of validation loss for 2.6 times the trainable parameters, but on LAMBADA, HellaSwag and PIQA the frozen-table model is better or tied; only ARC-Easy favours the learned ta...

• 2.Where Prefill Expert Reduction Hurts MoE Models: A Few Prompt Tokens Carry Most of the Short-Answer F1 Loss

  arXiv:2610.22454v1 Announce Type: new Abstract: Running prompt tokens through fewer routed experts during prefill is a simple way to make mixture-of-experts (MoE) serving cheaper, yet it is unclear what this damages. We locate the damage with token-, layer- and attention-path interventions on short-answer question answering in four MoE families. At half the experts the F1 loss of three families is mostly answer form: the model still finds the answer but stops answering briefly. In Qwen3-30B-A3B this loss is made at the two chat-template tokens before the answer, and in DeepSeek-V2-Lite at the first 8-32 prompt tokens, where its instruction sits; by answer likelihood these positions account for about half and a fifth of the damage. Reduction changes the first token's attention mass, norm and largest activation by less than 4%, and restori...

• 3.From Parallels to Divergence: Language Models' Learning of Action Nominal Constructions

  arXiv:2610.22468v1 Announce Type: new Abstract: Action nominal constructions (ANCs) express events in nominal form, with their argument marking following patterns found in clauses or ordinary noun phrases. In this paper, we investigate how different grammatical configurations affect language models' learning of ANCs and related constructions, and how these learning preferences compare with typological distributions in natural languages. We use a grammar-engineered method for synthetic corpus construction that combines shared semantic input from natural corpora with computational grammars. We construct 96 target languages varying in clause word order, noun-phrase word order, alignment, complement system, and ANC strategy. For each language, we train GPT-2-small models from scratch and evaluate them using targeted minimal pairs covering 34...

• 4.RealCLI: Toward a Unified Evaluation of Human-Agent Collaboration for CLI Agents

  arXiv:2610.22507v1 Announce Type: new Abstract: Command-line interface (CLI) agents increasingly execute longer-horizon tasks, yet existing benchmarks either assume that user instructions remain fixed throughout execution or evaluate task outcomes without capturing runtime collaboration quality. In this paper, we present REALCLI, a paired benchmark that unifies the evaluation of task execution and human-agent collaboration quality on realistic CLI tasks. To comprehensively measure collaboration, REALCLI builds a five-dimensional framework, operationalized into 22 sub-dimensions and 51 interaction scenario templates. REALCLI provides two modes for paired comparison: REALCLI-EXEC supplies complete specifications upfront, whereas REALCLI-COLLAB introduces a scripted user that delivers progressive information, corrections, constraint revisio...

• 5.AuditKIT: A Unified Evaluation Spine for Benchmarking, Judging, and Comparing Large Language Models and Agentic Systems

  arXiv:2610.22511v1 Announce Type: new Abstract: Evaluating language models and agents requires combining benchmark accuracy, judge-based assessment, runtime performance, tool-use behavior, and retrieval quality. Differences in interfaces, metric semantics, and trace formats complicate these comparisons and can hide failures in agentic workflows. We present AuditKIT, a source-available evaluation library that records these measurements in one result representation, with direction-aware metric contracts, configuration fingerprints, and extensible metrics and backends. For agents, it provides turn-aware tool-call matching, separates parallel from sequential execution, and distinguishes verified task outcomes from trajectory-level diagnostics. Missing or unobservable evidence is marked unscored rather than counted as failure. For retrieval-a...

• 1.When is a large language model a good calibration agent? Physics-labeled prompting for budgeted engineering simulator calibration

  arXiv:2610.22379v1 Announce Type: new Abstract: Calibrating expensive engineering simulators requires tuning model parameters to match reference data, a process heavily constrained by computational costs. This study investigates whether Large Language Models (LLMs), conditioned on physical diagnostic descriptions of simulation errors, can propose informative parameter updates more efficiently than conventional derivative-free optimizers. We test this on a cylinder-wake Large-Eddy Simulation and a damped-oscillator inverse problem. Physics-labeled prompts are evaluated against four text ablations (removed, anonymized, padded, misdirected) and standard optimizers (Nelder-Mead, Bayesian optimization, CMA-ES, and random search). Results demonstrate a strong early semantic advantage: the first physics-informed LLM proposal improves the initial...

• 2.Trust but Verify: Designing Region-Inhomogeneous Forward Dynamics for Discrete Diffusion

  arXiv:2610.22380v1 Announce Type: new Abstract: Continuous diffusion treats the forward process as a design space; discrete language diffusion often uses absorbing corruption, leaving context fully trusted throughout generation. This makes trust binary: corrupted context is copied as ground truth, propagating errors into generated spans. We reopen this design space by combining absorbing and uniform-mixing kernels with region-dependent rates. Missing positions follow absorbing dynamics, while context undergoes low-rate random substitution, exposing the model to observed but uncertain tokens during training. At inference, a heuristic sampler re-injects noise into suspicious context under an annealed intensity and freezes context updates late in the trajectory. An 8,000-step finetune of a pretrained 170M-parameter masked diffusion model tak...

• 4.CGDT-CESN: Correlation-Guided and Dynamics-Trained Clustered Echo State Network for Multivariate Time-Series Prediction

  arXiv:2610.22430v1 Announce Type: new Abstract: Accurate prediction of multivariate time series requires models that simultaneously capture variable-specific dynamics and cross-variable interactions. In this work, a Correlation-Guided and Dynamics-Trained Clustered Echo State Network (CGDT-CESN) is proposed, which comprises a modular reservoir architecture with each cluster distinctly specializing in one of the variables in the multivariate time series. An innovative reservoir design is proposed in which inter-cluster connections are enabled by empirically observed correlations among variables, providing a data-driven coupling mechanism. Apart from conventional readout training, inter-cluster connection weights informed by correlations are trained, which in turn allows the reservoir to capture time-varying cross-variable dependencies with...

• 1.DriveWorld3D Evidence Constrained 3D World Modeling and Generative Simulation for Real-World Autonomous Driving

  arXiv:2610.22450v1 Announce Type: new Abstract: Three dimensional modeling for autonomous driving must continuously estimate road geometry, obstacle occupancy, traffic participant behavior, and observational uncertainty under limited computational resources. Existing reconstruction methods can synthesize realistic novel views, but visual fidelity does not sufficiently characterize the geometric reliability required for driving. Generative world models can expand scenarios and predict future observations, yet may introduce content unsupported by measurements. We propose DriveWorld3D, a framework combining temporal video learning, measurement constraints from multiple sensors, and offline generative simulation. Its online component maintains a unified world state comprising road surfaces, volumetric occupancy, dynamic entities, and road rel...

• 2.Mothership: A Two Tier Autonomous Refuse Collection System for Pedestrian Campuses

  arXiv:2610.22461v1 Announce Type: new Abstract: Refuse collection in pedestrian campuses, parks and mixed-use districts remains manual because compaction trucks cannot reach bins located on walkways, courtyards and building service doors. This paper presents Mothership, a two-tier autonomous refuse collection system in which an electric compaction truck (the mothership) carries two low-speed sidewalk robots (RoboCarts) that fetch standard two-wheel bins, deliver them to the truck, return the empty bins to their original positions and re-dock on the truck for charging. The architecture separates responsibilities by scale: the truck performs road driving, compaction, lifting and energy supply, while the carts perform only sidewalk transport and bin grasping. This split removes the lift-and-tip mechanism from the sidewalk vehicle, keeping it...

• 3.Has GPT-6-Astra Achieved Embodied Intelligence?

  arXiv:2610.22471v1 Announce Type: new Abstract: GPT-6-Astra demonstrates a level of competence in robotic manipulation and navigation that challenges the conventional paradigm of building specialized models for individual tasks and embodiments. In this report, we ask not only how capable GPT-6-Astra is, but what its capabilities imply for the field. First, we probe the limits of standalone GPT-6-Astra. We conduct a large-scale evaluation across 12 representative benchmarks spanning tabletop manipulation, humanoids, mobile platforms, and dexterous hands. Under a unified protocol that allows in-context learning (ICL) from training-set demonstrations and substantially increases the number of execution steps, GPT-6-Astra achieves unprecedented performance (Fig. 1). Future research may focus on improving 1) the generality of ICL 2) cost and ex...

• 4.Scaling Online Exploration for Robot Agents

  arXiv:2610.22496v1 Announce Type: new Abstract: Robot learning has advanced by scaling demonstration datasets and model capacity; however, deploying a new robot embodiment or task typically necessitates a dedicated teleoperation corpus. The resulting policies remain bounded by their imitation distributions, limiting generalization and continual self-improvement. We introduce RAX (Robot Agent eXploration), an open-source framework for robot agents centered on scaling exploration: enabling agents to improve by acquiring online experience and synthesizing executable programs. In RAX, a general-purpose coding agent generates robot control code, operating without action demonstration data, hand-engineered action primitives, or value-function guidance. Through online exploration, the agent identifies decision sequences that achieve task goals a...

• 5.RTISSLAM-3D: Full 6-DoF SLAM With an In-Air 3D Imaging Sonar

  arXiv:2610.22552v1 Announce Type: new Abstract: Echolocating bats navigate dense forests using sonar alone, and in-air 3D imaging sonars now capture the entire frontal hemisphere with every pulse. Sonar SLAM, however, has so far been planar: current systems estimate a 2D trajectory and discard the elevation information in the sonar images, an assumption that does not hold on rough and sloped terrain such as a forest floor. In this paper, we introduce RTISSLAM-3D, a novel 6-DoF SLAM system for a wheeled robot that carries an in-air 3D imaging sonar, an IMU and wheel encoders. A CFAR detector on the sphere reduces every sonar image to a sparse set of 3D detections, which we use for a 3D acoustic odometry that models the shape of the sonar beam, for place recognition with polar submap descriptors, and for the verification of loop closures by...

• 1.Reading the News Without Hindsight: NoLBERT Topics and the Cross-Section of Stock Returns

  Language models that read financial news are usually trained on text from the same years used to test them, so a backtest can reward hindsight. We build news topics from NoLBERT, an encoder pretrained only on text published between 1976 and 1995, adapted into a sentence encoder by contrastive fine-tuning on passages of the same article, with no labels or returns. We compare it with LDA, MiniLM and FinBERT readers on 1.26 million articles from 2012 to 2024, convert each reader's topics into attention shocks and stock exposures, and forecast the returns of the 1,000 largest US stocks with eight models at three rebalancing frequencies. Models are fitted on 2013-2019, choices are made on 2020-2021, and 2022-2024 is held out. The experiment with the best validation score loses money at every frequency. Measuring attention shocks against the pa...

• 2.OrderFlowDiT: Transport-Regularized Diffusion Transformer for Probabilistic Limit-Order Supply Forecasting

  We study the conditional distribution of limit-order volume submitted over the next five minutes, given the current order book and recent order flow. The task requires capturing price placement, total volume, and predictive uncertainty jointly. Coordinate-wise diffusion losses do not explicitly account for price displacement, while normalizing fields to unit mass removes information about total volume. Agreement on selected summary statistics also does not establish fidelity to the joint field distribution. We introduce OrderFlowDiT, a conditional Diffusion Transformer for unnormalized bid and ask volume-at-price fields, trained with an unbalanced optimal-transport (UOT) auxiliary loss. We also develop a full-field evaluation framework that combines a transport cost sensitive to price placement and total volume with proper scoring rules a...

• 3.Momentum, Commentary Tone, and Earnings Surprise in LLM Financial Forecasts

  Which inputs drive a language model's financial forecast? We construct synthetic earnings announcements that independently reverse 20-day momentum, non-informational commentary tone, and whether fixed actual EPS beats a manipulated consensus. Six instruction/chat checkpoint conversions from four families (0.49-7.61B base parameters) receive 2,016 prompts each under a direct-answer Q8_0 protocol. The design includes 24 numerical blocks, six commentary pairs, profits and losses, rising and falling annual EPS, and two surprise sizes. Both A/B answer assignments and U/D cover all 384 primary scenarios per conversion. We estimate matched changes in conditional answer-prefix scores, not calibrated market probabilities. Across these checkpoints, the primary EPS contrast ranges from -0.19 to 86.80 percentage points. The magnitude of the primary s...

• 4.When Does "Post-Call" Begin? An Availability Contract for Earnings-Call NLP Benchmarks

  Corporate financial disclosures often unfold over time, with announcements, prepared remarks, and discussion spanning multiple high-frequency return intervals. Financial NLP benchmarks can therefore suffer from a within-event timing mismatch when the prediction target begins before the textual or audio input used by a model could be complete. Chronological train--test splits do not prevent this mismatch, because it occurs inside individual events rather than across samples. We introduce an Availability Contract, which requires a benchmark to state the input boundary, evidence for input availability, the outcome start, and the claims supported by that evidence. We apply this framework to 552 earnings calls from 24 firms. For each call, we hold the return endpoint fixed and compare the previous close, a documentary call-start proxy, and a c...

• 5.StochLOB: Heavy-Tailed Score Matching for Robust Limit Order Book Trend Prediction

  Score-based diffusion methods usually perturb data with Brownian Gaussian noise, corresponding to a light-tailed forward stochastic differential equation (SDE). For market time series this SDE can be a poor abstraction: limit order book (LOB) order flow is heavy-tailed, bursty, and occasionally jump-like. We study STOCHLOB, a training-only score-matching framework that changes the forward diffusion corruption law while leaving the deployed predictor deterministic. Instead of Brownian-only Gaussian corruption, the noising process uses two non-Gaussian SDE-inspired alternatives: jump-Lévy noise through a compound-Poisson variance mixture, and clipped $α$-stable noise through a positive-stable Gaussian scale mixture. These corruptions inject finite jumps and heavy-tailed residuals into training without requiring reverse-time diffusion sampli...

• 1.GSMA: Digital Reforms Could Connect 1.12 million More People in Madagascar and Add MGA 4.1 trillion to GDP by 2031

  Summary available at source link.

• 2.1.2 billion people across Developing Asia Pacific remain offline despite near universal mobile coverage

  Summary available at source link.

• 3.GSMA Board Sets Out Global Priorities for the Next Era of Connectivity

  Summary available at source link.

• 1.Beyond Large Language Models: A Mental World Model Framework for Mental Modeling, Simulation, Evolution, and Interpretability

  Understanding and simulating how subjective cognition and mental states emerge, evolve, and shape behavior is essential for advancing artificial intelligence toward deeper social cognition and general intelligence. Existing large language models (LLMs) and Theory of Mind (ToM) methods primarily infer mental states from linguistic information, lacking unified mechanisms for subjective world construction, cognitive-psychological interactions, dynamic evolution, and historical dependencies. This paper introduces the Mental World Model (MWM) and its implementation, the Multi-Branch Mental Evolution Architecture (MMEA), which explicitly constructs subjective worlds, models interactions between cognition and mental states, and simulates continuous mental evolution shaped by historical experiences. On DynToM and RecToM, MMEA achieves 96.48% and ...

• 2.From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents

  Large language model (LLM) agents tackle complex tasks through reasoning and interaction with external environments. Agent skills provide persistent, transferable capabilities that guide task planning, inform contextual decisions, support tool use, and enable executable procedures. Yet existing skill-evolution methods largely organize knowledge around skill artifacts, overlooking the functional heterogeneity within individual skills and its implications for knowledge maintenance and use. Inspired by cognitive science, we propose Cognitive Differentiation in Skill Evolution (CDSE), which evolves a unified skill through four functionally distinct yet coordinated components: Meta, Experience, Tool, and Script. CDSE extracts evidence-grounded semantic gradients from successful and failed trajectories, consolidates them by functional role, and...

• 4.SkillBrowser: Selective Skill Exposure for Scalable Skill-Using Agents

  Large language model agents increasingly rely on skills to extend their capabilities in complex tasks. However, scalable skill use requires Selective Skill Exposure (SSE): providing relevant procedural content when needed while avoiding unnecessary exposure that may disrupt reasoning. Existing paradigms fall short in complementary ways. Progressive disclosure exposes metadata for a growing skill library, introducing context pollution; retrieval limits catalog exposure, but static retrieval from the initial task query may be misaligned with capability needs that emerge during reasoning and may still inject unnecessary skills. We propose SkillBrowser, a Meta-Skill-managed framework that enables an agent to identify current capability needs, formulate targeted queries, and selectively load skills on demand. Across six SRA-Bench subsets and f...

• 5.Trapped in Your Story: Evaluating Narrative-Induced Sycophancy in Large Language Models

  Large language models (LLMs) are known to exhibit sycophancy by adjusting their responses toward users' expressed views. However, real users often justify their opinions through reasons, personal experiences, causal explanations, and broader narratives, raising the question of whether such narratives can further steer model judgments. We construct a controlled benchmark of 500 debatable questions across ten domains, where the user stance is held fixed while its justification is systematically varied without introducing new task-relevant evidence. Each question is instantiated under 11 matched conditions: one neutral baseline and five mirrored narrative levels for each of two opposing positions, yielding 5,500 evaluation instances. We evaluate seven LLMs and observe substantial differences in narrative-induced agreement. The effect is not ...

• 1.Forecasting Voltage Time Series under Anomalies Using a WGAN Model With Lipschitz Control and Progressive Kernel Critic Strategy

  In time-series forecasting tasks, the use of recursive forecasting on high-dimensional data substantially increases the complexity of the prediction process. The main challenge is that long-term dynamics becomes harder to model, since the sequences show greater temporal variability. To counteract this limitation and improve the modeling of long-term dynamics, and more importantly, address high-dimensional recursive forecasting, this paper uses a Wasserstein GAN equipped with a progressive Lipschitz penalty and an adaptive kernel. This model is capable of taking a window of the data stream and forecasting the next values, subsequently reconstructing the time-series stream and dynamically adapting its kernel to each new window. The proopsed model is a deep learning framework based on a Wasserstein GAN (WGAN-CLP) with an adaptive kernel to m...

• 2.AgentOnto: Ontology-Guided LLM Agents for Enterprise Data Analytics

  Enterprise data analysis requires agents to connect analytical requests with knowledge distributed across databases, business documents, and existing example tasks. Without an explicit representation of this knowledge, an agent may produce incorrect analytical results and cannot reliably incorporate feedback. We present AgentOnto, an ontology-guided LLM agent for enterprise data analytics. AgentOnto organizes heterogeneous enterprise knowledge into an ontology that directly guides analysis and incorporates confirmed feedback. Iterative Ontology Preparation further improves the ontology through self-generated analysis tasks. We evaluate AgentOnto on two public data sources. AgentOnto substantially outperforms representative methods, while module-level evaluations confirm the effectiveness of ontology construction, analysis, and updating. A...

• 3.Path to Prior-Data Fitted Relational Models via Learned Latent Multi-table Representations

  Prior-data fitted networks (PFNs) show that pretraining on large populations of synthetic tasks can enable generalization to unseen datasets with little or no downstream fine-tuning. This paradigm is effective for transformer-based tabular and relational foundation models such as RDB-PFN, but remains largely unexplored for graph-based relational models built on message passing. We pretrain the graph-based relational foundation model Griffin on large collections of synthetic relational databases and ask whether curriculum learning, which benefits transformer-based relational learners, also improves message-passing representations. Rather than fine-tune on downstream tasks, we use the pretrained encoder as a general-purpose relational representation learner and pass its latent representations to a frozen in-context predictor (\ticl{}); no c...

• 4.ASE: Attribution-Guided Semantic Evolution for Time Series Forecasting

  External text can supply predictive information absent from numerical histories, but its value depends on how it is represented, combined and deployed. We present Attribution-guided Semantic Evolution (ASE), a modular framework for constructing textual factors, evaluating their conditional contribution, and deploying validation-selected numerical--textual predictors. ASE supports task-specific implementations with lexical measurements, semantic primitives or evolved event/state factors, coupled to suitable numerical experts. Its feedback-based implementation uses conditional refitting diagnostics to guide factor revision and combination search. We evaluate ASE on the four text-associated TESS datasets and dynamic CSI 300 stock ranking. The four forecasting implementations obtain test MSEs of 0.001041, 0.022439, 0.333115 and 1.8150, each b...

• 5.BRICE: Brain-inspired Recurrent Illusory Contour Encoder for Point Cloud Completion

  Mainstream point cloud completion methods are unanimously based on deep learning architectures. On one hand, these methods tend to repetitively stacking attention modules to enhance network performance, which drifts away from the neat design of unified and interpret-able networks; on the other hand, nearly all existing architectures rely solely on feed-forward connections to propagate information, neglecting feedback from deeper layers. The recurrent neural architecture, formed by feedforward-feedback loops, have shown its ability to generate Illusory Contours (ICs) in the V1 brain region of mammals; and ICs are essentially a completion mechanism of the brain for visual information. Motivated by this, we design a Brain-inspired Recurrent Illusory Contour Encoder (BRICE) to "imagine" what is missing in the incomplete 3D point clouds. Build...

• 1.Edge AI for Smart Cities: From Insight to Cognitive Action

  This white paper explains how edge AI and physical AI help cities move beyond passive monitoring. It shows how urban infrastructure can sense, reason, and act on real-time conditions across public safety, mobility, and infrastructure management. What you will learn about: How the shift from smart cities to cognitive cities depends on infrastructure that reasons and acts, rather than infrastructure that only monitors. Why edge AI supports low-latency decisions, keeps sensitive data within city boundaries, and reduces dependency on centralized cloud infrastructure. How computer vision, multimodal sensing, and generative and agentic systems work together to support public safety, mobility, and infrastructure applications. How a tiered edge architecture assigns monitoring, planning, and action tasks across the far edge, regional, and cloud la...

• 2.Simulation App Speeds Up Electric Truck Battery Design

  This article is brought to you by COMSOL . These days, no one blinks an eye at two-day, one-day, and even same-day package deliveries. Despite the plethora of localized fulfillment and distributi on centers that have cropped up to meet the demand, delivery trucks still need to make considerably long drives, often over highways. To reduce fuel and maintenance costs as well as carbon emissions, companies are investing in electric trucks for their delivery fleets. While turning to electric trucks is a promising solution, the battery packs incorporated into these automotive designs often fall short in the lifetime needed to make long-distance highway travel feasible. To help overcome the issue of battery lifetime, a consortium of universities and industry partners incorporated multiscale simulation and a custom simulation app into the battery...

• 3.Google DeepMind’s New Resident Philosopher Tackles AI Ethics

  How should humans relate to generative AI ? In what ways do we, or should we, treat a large language model (LLM) as if it were a person? If an AI became conscious, how could we tell? Does it make sense to think AI could someday have moral rights? These are among the cognitive and ethical questions that occupied philosopher Henry Shevlin at Cambridge University’s Leverhulme Centre for the Future of Intelligence . But last May, Shevlin became part of the growing cohort of philosophers who work inside leading artificial intelligence companies. Though he remains active at Cambridge, he now spends 80 percent of his time at Google DeepMind (job title: “Philosopher”). Shevlin spoke with IEEE Spectrum about why philosophers are drawn to AI, how humans and AIs relate to each other, fears of “rogue” AI and AI relationships that harm humans, and how...

• 1.The California billionaire tax is very likely a net revenue loser

  This paper examines the revenue and migration impact of the 2026 California Billionaire Tax Act, a one-time 5 percent tax on the worldwide net worth of individuals exceeding $1 billion. Using data from the Forbes Billionaire List, news reports detailing the residential real estate holdings of California billionaires, and public announcements of billionaire departures from […]

  The post The California billionaire tax is very likely a net revenue loser appeared first on Marginal REVOLUTION.

• 2.No, DataCenters are Not Crowding-Out Other Investments

  Based on charts like the one at right, Paul Krugman and many others are arguing that: Massive spending on data centers is crowding out investment in everything else in the economy. Not so fast. Can you see one immediate problem? How is $51b in data center construction spending crowding out $120b in everything else! Indeed, […]

  The post No, DataCenters are Not Crowding-Out Other Investments appeared first on Marginal REVOLUTION.

• 3.What I’ve been reading

  David Stuart, The Four Heavens: A New History of the Ancient Maya.  The best book about the ancient Maya that I know of, and also a book about how we keep on discovering more about them.  “The model I advocate for here rejects the old idea of the rise and fall of the Maya in […]

  The post What I’ve been reading appeared first on Marginal REVOLUTION.

• 4.*You Can See Everything* (no spoilers)

  An excellent and also original movie, you all should see it.  There are different interpretations, but I read it as a story between two autistic people, and the bond they have but also the great difficulties they have in understanding each other, because they are so very different.  I am not saying that is what […]

  The post *You Can See Everything* (no spoilers) appeared first on Marginal REVOLUTION.

• 1.Revenue per Worker Is Up. Are Firms Getting a Larger Share?

  From the Great Recession until 2019, U.S. labor productivity growth disappointed. Nonfarm-business output per hour grew about 1 percent a year from 2010 to 2019, down from 3 percent in the late 1990s. This slowdown was especially pronounced in manufacturing, and it coincided with a slowdown in the growth of revenue per worker among public firms. However, since 2019, growth has picked up to twice the 2010s rate, around 2 percent. In this post, I use data from firms’ 10-K filings to unpack this recent acceleration. I first show that the acceleration appears in public firms’ ...

• 2.Has the AI Boom Impacted the U.S. Current Account?

  The AI boom is showing up in different corners of the U.S. economy, including international trade. Imports of semiconductors, servers, and computing equipment have climbed sharply yet the U.S. current account deficit has been largely unchanged. To better understand how the AI boom has not substantially affected the trade balance, consider that the current account is dictated by the gap between domestic saving and investment spending. It turns out that aggregate saving and investment spending have not changed much during the AI boom. Aggregate investment has not risen due to weakness outside the AI sector and business saving has remained robust.

• 1.PS Events: Reinventing Europe

  Summary available at source link.

• 2.A Nobel for the Poor People’s Lawyer

  From growing up in a slum outside Durban to studying law in a segregated university, Navi Pillay’s experience in apartheid-era South Africa instilled in her a lifelong belief in the importance of international courts in defending people’s rights. Now she should use her newly won stature to advance the rule of law at home.

• 3.MAGA at the Acropolis

  US Secretary of State Marco Rubio’s ideologically loaded recent speech in Athens attempted to trace MAGA’s ideological roots back to ancient democracy. But emphasizing the West’s shared civilizational inheritance will not convince Europeans to embrace the hubris of the Trump administration’s imperial worldview.

• 4.Breathing New Life Into Organized Labor

  Fulfilling the International Labour Organization’s promise of defending workers’ rights will require more than another declaration or incremental institutional reform. Instead, there must be a concerted global effort to create an enabling environment for all workers and their organizations to operate across borders as easily as capital does.

• 5.The State That Remembers

  Although many economists believe that AI could drive an economic transformation larger than the Industrial Revolution, and on a vastly shorter timetable, most governments are unequipped to manage such a transition. We urgently need states that can learn on the go and remember what they learn.

• 1.Regulators look to WRC-27 to settle direct-to-device spectrum rules

  Spain, U.K. and U.S. regulators map different roads to WRC-27 In sum – what we know: Direct-to-device connectivity has run ahead of the international rulebook. Reusing spectrum allocated solely for terrestrial…

• 2.US telecom retail holds up – where might Starlink fit? (Analyst Angle)

  US telecom retail remains resilient, but Wave7 Research’s Jeff Moore questions whether Starlink can become a national mobile operator. SpaceX could bypass the traditional retail model, but building a competitive…

• 3.AI in the RAN moves toward the scale test

  Speaking at the RCR Wireless News’s Telco AI Forum 2026, AT&T and Verizon executives described AI-based RAN deployments that are already producing measurable results In sum – what to know…

• 4.Starlink’s MO – replacement or displacement? (Telco Diary)

  From the newsletter (October 16; sign-up or follow if you want it sooner): Starlink need not replace terrestrial networks to disrupt telecoms. Strand Consult and Dell’Oro examine SpaceX’s hybrid ambitions,…

• 5.6G architecture is looking like 5G made smarter, not a clean slate

  Panel maps the 6G path from standalone 5G to agentic AI cores In sum – what we know: 6G has started moving from vision to engineering, and with it comes…

• 1.A Smaller Subspace is All You Need: Efficient ISAC Beamforming via Dimensionality Reduction

  Integrated sensing and communication (ISAC) has been recognized as a critical technology for the sixth generation (6G) wireless networks. Under monostatic sensing mode, achieving a balance between communication and sensing performance via optimal beamforming is a critical research problem. In multiple input multiple output (MIMO) systems, such beamforming problems are of substantial computational complexity due to the large number of antennas. Existing methods exploit exact subspaces of lower dimensions to speed up the beamforming process where the complexity is proportional to the number of communication users and sensing targets instead of antennas. In this paper, it is demonstrated that the beamforming problem can be solved in an even lower-dimensional subspace with only a small sensing performance loss. Numerical results compare the s...

• 2.LWMa: Distributed Deployment of an Angle-Delay-Domain Wireless Foundation Model for Multiple Tasks

  The growing diversity of wireless services in sixth-generation (6G) wireless communication systems calls for reusable channel representations that support multiple tasks across different propagation environments. However, deploying such representations between user equipment and the base station in frequency-division duplexing systems faces uplink feedback constraints. This paper proposes a distributed framework based on an angle-delay-domain large wireless model (LWMa) for multiple tasks. LWMa exploits angle-delay structure and delay truncation to improve feature extraction and reduce computational cost. A scenario-specific network inspired by Perceiver IO (input-output) compresses embeddings for channel reconstruction tasks, while other tasks use a low-dimensional compact global embedding. Embedding reconstruction training followed by c...

• 3.Cryptographically Secure Agent-to-Agent Communication in Multi-Agent Systems

  Multi-agent systems (MAS) comprise AI agents that collaborate or compete to perform complex, distributed tasks. Their dynamic interactions create opportunities for threat actors to target and impersonate other agents, intercept messages, inject malicious prompts, and exfiltrate sensitive data while simultaneously complicating the detection of anomalous behavior through conventional monitoring. This paper surveys threats and cryptographic challenges in agent-to-agent communication and presents a zero-trust architecture integrating identity, transport security, message integrity, and monitoring. The architecture establishes agent identities via decentralized identifiers and task-scoped verifiable credentials backed by a permissioned ledger, enforcing least-privilege authorization. For transport security, it provides mutually authenticated, ...

• 4.Integrated Sensing and Communication in 3GPP: Standardization, Performance Evaluation, and Experimental Validation Toward 6G

  Integrated sensing and communication (ISAC) is evolving from a primarily physical-layer research into a standardized cellular-network capability spanning channel modeling, radio procedures, network architecture, and the delivery of sensing services. We review this progression within the Third Generation Partnership Project (3GPP), from Release 19 service requirements and target-aware channel modeling to the Release 20 New Radio (NR) sensing evaluation framework, radio access network procedures, and system architecture, and discuss future directions toward 6G. We then present two 3GPP-aligned simulation case studies to demonstrate the corresponding evaluation approaches. The first case study compares hard-decision, soft-decision, and hybrid fusion for unmanned aerial vehicle (UAV) sensing using multiple transmission and reception points (T...

• 5.LLM-Guided Dual-Stage Latency Optimization for Closed-Loop 6G V2I Digital Twins

  Wireless Digital Twins are emerging as a key enabler for predictive network control in 6G Vehicle-to-Infrastructure (V2I) systems. However, closed-loop V2I DTs face two dual-stage communication bottlenecks arising from model parameter exchange for V2I DT reconstruction and multi-service V2I traffic delivery after each reconstruction. This paper proposes a closed-loop dual-stage End-to-End (E2E) latency optimization framework that integrates Compressed Personalized Federated Learning with a unified Large-Language-Model-guided Multi-Agent Reinforcement Learning approach. A CPFL-enabled decision-oriented V2I DT is first constructed by jointly predicting Path Loss (PL) and Quality-of-Experience (QoE)-oriented Key Performance Indicators (KPIs). Then the framework comprises two temporally coupled but latency-wise distinct optimization stages: S...

• 1.Market Efficiency in a Changing Asset Population: Mobility, Entry, and Survival over Cryptocurrency's Full History

  Weak-form efficiency implies symmetry restrictions on the Markov chain governing movement across the cryptocurrency capitalisation distribution. We test them on the full listed population: 8,970 assets, 690 weekly cross-sections, 2013 to 2026, with entry, exit and survival observed rather than inferred. Symmetry is rejected in every era, but the asymmetry falls by seventy percent, and what converged is entry-driven compositional churn rather than coins' own valuations. The bottom decile earns 27.5% per week on paper and nothing among assets that trade, on capacity of $293,000. Compensation for neglect exists precisely where trading does not.

• 4.Where Backtests and Live Trading Part Ways: Six Defects from One Intraday System

  Backtests of intraday strategies can drift away from what the live system actually does; in our case, the drift never showed in summary statistics. We report six defects found while running one intraday US equity system live and comparing its live days with their backtest replay. Three came from data: a default feed that carried a small fraction of real opening volume, historical prices adjusted for dividends that live orders never see, and data the live system downloaded but the backtest never did. Two came from code: a sizing step that exceeded a limit the live account enforces, and UTC timestamps read as New York time. One came from configuration: settings that were switched on but never reached the code that places orders. For each we give the symptom, the size of the error, the cause and the test or control that now guards against it...

• 5.Pricing the Unprovable: Ambiguity, Turnstile Runs, and the Verifiability Discount in Crypto Protocols

  Privacy-preserving protocols can repair a vulnerability without establishing whether it was exploited beforehand, even when no theft is reported and aggregate supply is protected. We call the potential value reduction from this gap the verifiability discount. A turnstile model shows how early withdrawal can shift a possible shortfall to remaining holders and why credible evidence can become more valuable as exposure grows. We introduce VDISC, a reproducible framework that records reported loss, exposed value, public evidence, returns, and pool flows separately. It combines Orchard's staged disclosures with reconstructed on-chain flows, earlier Zcash cases, and a 40-event security panel. Orchard follows the predicted direction: market responses were muted after repair and supply assurance, while the largest negative return and aggregate sh...

• 4.ScaleRAN: A Runtime for Elastic GPU Resource Management in AI-RAN

  AI-RAN is driving the convergence of radio access network (RAN) processing and artificial intelligence (AI) workloads on GPU-enabled computing platforms, ranging from GPU-accelerated base stations to cloud-native infrastructures. While GPU acceleration improves the efficiency of virtualized RAN functions, the computational demand of latency-critical baseband processing varies with traffic load, creating transient spare GPU capacity that can be reclaimed by co-located AI applications. Existing GPU-sharing mechanisms rely on coarse-grained hardware partitioning or static resource allocation, limiting their ability to adapt to slot-level workload fluctuations while preserving strict timing requirements. This paper presents ScaleRAN, a node-local runtime for elastic GPU resource management in AI-RAN. ScaleRAN leverages NVIDIA Green Contexts...

• 3.Towards Virtual Networking for a Time-Triggered Hypervisor

  Virtualization enables mixed-criticality applications to run on shared hardware by isolating them within separate virtual machines (VMs). However, achieving end-to-end timing guarantees for critical tasks hosted on virtual CPUs (VCPUs) without relying on static partitioning is hard. A possible solution is a time-triggered hypervisor that dispatches VMs' VCPUs to processor cores according to a predetermined global schedule. A global schedule includes the computation, the network, and the virtualization layer to provide end-to-end timing guarantees across hosts and VMs during design time. In this work, we propose a VirtIO-based virtual network that realizes IEEE 802.1Qbv quasi-compliant inter-VM communication, providing predictable timing for critical messages exchanged between time-triggered VMs. We present our VirtIO-based virtual network...

• 4.QoE-Aware Service Function Chain Orchestration in Space-Air-Ground Integrated Networks for In-Flight Internet

  In remote regions lacking terrestrial infrastructure, aviation users increasingly demand Internet services comparable to ground cellular networks. This paper proposes a space-air-ground integrated network (SAGIN) architecture that combines ground networks, civil aviation networks, and low Earth orbit (LEO) satellites. Based on real North Atlantic flight data, a heterogeneous aviation Internet is constructed to support diverse aircraft service requirements. A composite Quality of Experience (QoE) metric, characterized by transmission delay, jitter, and content similarity, guides efficient resource utilization. The dynamic connectivity of SAGIN is modeled using a reconfigurable time-expanded graph (RTEG). With Network Function Virtualization (NFV), data delivery is formulated as a Service Function Chain (SFC) deployment across ground stat...