The Morning Preprint

Thursday 3 September

Showing the most recent batch — Thursday 3 September. No papers have been ingested today yet.

666 papers ingested by the estate's stations that day — about 3 minutes to read. The counts and the chart are the estate's own ingest, not arXiv's submission figures.

This day's harvest looks short: 666 papers against a busiest day of 1650 in the week shown below. Weekends and a broken fetch look the same in the chart, and the chart counts the estate's ingest, not arXiv's.

The 60-second version

Today's batch, and the 60-second version is still being written. Everything below is today's own work: the papers as they are landing, the stations that have written their piece, and the day's episodes. This page fills in as the day does.

The week in the estate's ingest

Seven days of papers the estate ingested. Weekends are striped — the weekend is visible in the data.

742Fri
643Mon
1650Tue
956Wed
666Thu

Show the answer

666 papers were ingested on Thursday 3 September.

666 papers, three ways

Where today's batch landed

By the paper's primary arXiv category.
cs.LG131
cs.CL119
cs.AI89
cs.CR42
cs.CV31
astro-ph.GA25
astro-ph.HE23
stat.ML20

The index

The day's strongest 60, by score. Every title links its paper here and on arXiv.

  1. Untangling the Mechanisms of Misleading Context in Medical Question Answering [arXiv] [episode]
  2. Constraining reionization-era Ly alpha escape with JELS-MUSE: a highly complete H alpha-selected sample at z about6.1 [arXiv] [episode]
  3. JWST Reveals a Candidate Supermassive Black Hole Binary at z=4.3 in the Brightest Sub-millimeter Galaxy in COSMOS-Web [arXiv]
  4. A Common Measure of Communication for Speech Brain-Computer Interfaces [arXiv] [episode]
  5. Differentiable Electricity-Market Clearing for Gradient-Based Planning [arXiv] [episode]
  6. Semantic Signal-Assisted Inspection and Recovery Allocation in Reverse Logistics [arXiv] [episode]
  7. Evidence for Shared Routing Geometry and Dynamics in Sparse Mixture-of-Experts [arXiv] [episode]
  8. DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents [arXiv] [episode]
  9. An updated census of the Vast Polar Structure member galaxies [arXiv] [episode]
  10. Multiplicity of Cool Stars and their Evolution [arXiv] [episode]
  11. Cliff: Learning Process Rewards from the First Mistake [arXiv] [episode]
  12. Learning Spectrally Optimised Mesh-Free Discretisations [arXiv] [episode]
  13. Not All Agreement Counts as Corroboration: Provenance-Conserving Multi-View Fusion for Typed Action Admission in Human-Robot Collaboration [arXiv] [episode]
  14. HyperMC: Multi-Fidelity Hyperparameter Tuning for Stochastic Gradient MCMC [arXiv] [episode]
  15. Hardware-Accelerated Instance Segmentation for Resource-Constrained Space Robotics with Criticality Analysis [arXiv] [episode]
  16. From orbit to ground: pre-impact meteorite strewn field predictions for imminent impactors and meteorite recovery [arXiv] [episode]
  17. How LLMs Build Fictional Worlds: Setting and Narrative Space in AI-Generated Creative Storytelling [arXiv] [episode]
  18. Overcoming the Randomness-Utility Trade-off in Answering Differentially Private Linear Queries [arXiv] [episode]
  19. Predict, Don't Iterate: Efficient Adaptive-Length Infilling for Diffusion Language Models [arXiv] [episode]
  20. Demonstrating the Time-Domain Capabilities of the 4-m International Liquid Mirror Telescope: An Early Census of Transient and Variable Detections [arXiv] [episode]
  21. Omega-N: Interpretable Structural Node Descriptors and Their Applicability Domain [arXiv] [episode]
  22. Disease Burden over Skin Tone: Decomposing the Dermatology-AI Generalization Gap [arXiv] [episode]
  23. Solar Soft X-ray Coronal Dimming in a Failed Eruption Associated with Plasma Cooling [arXiv] [episode]
  24. Post-Training Language Models for Gold-Medal Performance in Coding Competitions [arXiv] [episode]
  25. text2ql: Multi-Target Natural Language Querying via a Language-Agnostic Intermediate Representation [arXiv] [episode]
  26. Can Risk-Based Alerting Mitigate Cybersecurity Alert Fatigue? [arXiv] [episode]
  27. When Decodability Is Not Enough: Logical Validity Representations, Behavioral Dissociation, and Causal Tests in Language Models [arXiv] [episode]
  28. TaRA: Training-Aware Low-Rank Adaptation Initialization [arXiv] [episode]
  29. SoK: Where Do Flow Labels Come From? Auditing Label Provenance in Encrypted Traffic Benchmarks [arXiv] [episode]
  30. Cite or Decline: A Strict Course-Grounded Chatbot for STEM Lecture Videos [arXiv] [episode]
  31. Fair Stable Matching: A Nash Social Welfare Approach [arXiv] [episode]
  32. POSEIDON III: The Aligned Orbit of the Hot Neptune Around the Hot Star WASP-195 [arXiv] [episode]
  33. LoRA-TSD: Tangent-Space Spectral Descent for LoRA via Muon-Style Updates [arXiv] [episode]
  34. Predictors of Loneliness in Older Adults Using Multimodal Analysis of Speech and Language [arXiv] [episode]
  35. LHAASO-WCDA observed a about 5 days TeV-delayed flaring event in blazar 1ES 1959+650 [arXiv] [episode]
  36. PRISM: An Agentic Multi-Model Architecture for Proactive Safety in Autonomous Transportation Systems [arXiv] [episode]
  37. SocialBuddy: Tailoring Search Agent for Social Scenarios [arXiv] [episode]
  38. Chemical evolution of Na, Mg, and Al in the Galactic bulge from UVES data [arXiv]
  39. InKAN: B-Spline KANs via Truncated Power Form [arXiv]
  40. Network-Aware Forecasting on Wireless Access Points [arXiv]
  41. Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment [arXiv]
  42. A Unified Particle Filter LSTM for Data-Driven Process Simulation [arXiv]
  43. NS-Copilot: An LLM-Driven Agent System for Autonomous Neuroscience Analysis [arXiv]
  44. Knowing Is Not Enough: Information Retrievability as a Precondition to Effective LLM Oversight [arXiv]
  45. Inflationary Magnetogenesis with f(R, phi) Coupling [arXiv]
  46. Benchmarking Language Models for Statistical Problem Formulation [arXiv]
  47. When Agents Implement Systems: A Case Study in Defects, Detection, and Evaluation Rigor [arXiv]
  48. Morphology signal in whole slide image foundation models can automatically triage slides [arXiv]
  49. CAHR-Net: Condition-Adaptive Hysteresis Reconstruction for Compact and Interpretable Magnetic Core Loss Modeling [arXiv]
  50. ClaimReceipt: Verifying Evidence Sufficiency and Coverage in Agent Evaluations [arXiv]
  51. Linear Fusion MultiDiffusion for Fast Training-Free Spherical Panorama Generation [arXiv]
  52. Posterior Tempering Explains Variance Inflation in Linear and Generalized Linear Thompson Sampling [arXiv]
  53. InsightSeg: Reusing Correction Insights for Guideline-Consistent Segmentation [arXiv]
  54. InstEditSeg: Instruction-Driven Image Editing for Polyp and Skin Lesion Segmentation [arXiv]
  55. Train What You Deploy: Closing the MLP Reachability Gap in Low-Rank Clone Distillation [arXiv]
  56. C squared T-OpenMax: A Novel Open-Set WiFi RF Fingerprinting Method via Center Constrained Learning and Confidence-Guided Tail Modeling [arXiv]
  57. Seed-Anchored Budget-Bounded Graph Rendering for Question Answering on Industry-Standard Power-Grid Information and Exchange Models [arXiv]
  58. What Would it Cost to End Extreme Poverty? [arXiv]
  59. How Output Format Confounds Data Quality and Capability in Instruction Tuning [arXiv]
  60. Perceptually Regularized Diffusion Model for Image Super-Resolution [arXiv]

Listen

Everything the estate published on 2026-09-03, in the player. Playing from this page is how the edition learns whether anyone did.