The Morning Preprint

Wednesday 2 September

Showing the most recent batch — Wednesday 2 September. No papers have been ingested today yet.

956 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.

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.

739Thu
742Fri
643Mon
1650Tue
956Wed

Show the answer

956 papers were ingested on Wednesday 2 September.

956 papers, three ways

Where today's batch landed

By the paper's primary arXiv category.
cs.CL201
cs.LG171
cs.AI153
cs.CR62
cs.CV53
astro-ph.GA36
astro-ph.HE28
astro-ph.SR26

The index

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

  1. A Checklist to assess the energy and carbon impacts of ML/AI applications in Earth System Modeling [arXiv] [episode]
  2. Convergence issues in Relational Concept Analysis based on AOC-posets [arXiv] [episode]
  3. Bandits in Prod: Hyperparameter Optimization at Inference Time [arXiv] [episode]
  4. Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts [arXiv] [episode]
  5. Latent Recurrent Thoughts: Recurrent Refinement of Proposed Latents for Reasoning with Frozen LLMs [arXiv] [episode]
  6. Higgsino Dark Matter Interpretation of the LUX-ZEPLIN 248 keV Nuclear-Recoil Event [arXiv] [episode]
  7. QTEA: Ternary LLMs with Sparse Residual Salient Weight and By-Column Optimization [arXiv] [episode]
  8. Creative Generation via Multi-Agent Debate: Does Debate Suppress Diversity? [arXiv] [episode]
  9. Beyond Magnitude: Contrastive Routing for Modular Mixture-of-Experts [arXiv] [episode]
  10. A Formal Analysis of Agent Payment Protocols [arXiv] [episode]
  11. Different representation learning objectives recover distinct latent structures from the same psychometric data [arXiv] [episode]
  12. Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs [arXiv] [episode]
  13. LLM-as-a-Demographic: Whom Sociodemographic Prompting Helps, and Whom It Hurts [arXiv] [episode]
  14. Invalidation Contracts for Cross-Episode Agent Memory [arXiv] [episode]
  15. Authority Bias in Conversational Search Engines for Academic Paper Recommendation [arXiv] [episode]
  16. Geometry-aware Latent Autoregressive Generative Model for PDEs in Complex Domains [arXiv] [episode]
  17. RestoreBench: Can AI Agents Restore Power Flow Convergence? [arXiv] [episode]
  18. A convolutional framework for detecting event-driven dynamics in energy price series [arXiv] [episode]
  19. REAL-Q: E2E LLM Quantization via Dynamic Gradient Descent [arXiv] [episode]
  20. Socrates went Nuclear: Comparing Interaction Strategies for AI systems in a Learning Context using Brain Sensing [arXiv] [episode]
  21. Confess What You Know: Forget-Set Misalignment with Model Knowledge in LLM Unlearning [arXiv] [episode]
  22. Nanohertz Gravitational-Wave Constraints on Supermassive Binary Black Holes at Cosmic Dawn [arXiv]
  23. Benchmarking Vision-Language Models for Automated Pathology Diagnosis and Report Generation [arXiv] [episode]
  24. Optimizing Byzantine Node Placement in Decentralized Federated Learning [arXiv] [episode]
  25. Scientific Agent Skills: A Library of Procedural Knowledge for Research Agents [arXiv] [episode]
  26. Soft-Argmax for the Projective Plane via the Veronese Embedding [arXiv] [episode]
  27. Trust Your Guide Only When Certain: Uncertainty-Aware Sparse Alignment at Inference Time [arXiv] [episode]
  28. Local Reference Geometry Residual Augmentation for Imbalanced Time Series Classification [arXiv] [episode]
  29. The Formation of Very Close Binary Stars [arXiv]
  30. Investigating Hyperparameter Optimization and Transferability for ES-HyperNEAT: A TPE Approach [arXiv] [episode]
  31. Are Near-Tied LLM Rankings Robust to Family-DIF-Guided Benchmark Recomposition? [arXiv] [episode]
  32. A Version Space Approach for Digital Circuit Analysis [arXiv] [episode]
  33. PCoMoE: Shifting MoE Inference from Monolithic Expert Selection to Fine-Grained Path Composition [arXiv] [episode]
  34. IntroConformal: Conformal Factuality Guarantees for Large Vision-Language Models via Introspective Signals [arXiv] [episode]
  35. Stochastic complexity of vectors containing cluster structure [arXiv] [episode]
  36. Modeling Rotation in the Old, Cold Domain: Implications on Gyrochronology and the Stellar Magnetic Wind [arXiv]
  37. Conversation Coach: A Voice-enabled AI System that Helps Practice Difficult Workplace Conversations [arXiv] [episode]
  38. Transferable End-to-End Optimization for Indirect Long-Term Memory Poisoning in LLM Agents [arXiv] [episode]
  39. A Mathematical Framework for Legacy, Governance, and Decision Integrity in Enterprise AI [arXiv] [episode]
  40. DNC-IMM: Early Lane-Change Intention Recognition via Neural Calibration Based on Driving Context Information [arXiv] [episode]
  41. Operational Regimes in Non-Convex Optimization: A Multiplier-Based Taxonomy [arXiv] [episode]
  42. Life Operators: a self-evolving framework for multiscale life modelling [arXiv] [episode]
  43. Breaking the Structural Identity: Personalized Federated LoRA Fine-tuning under Rank Heterogeneity [arXiv] [episode]
  44. From Base Rollouts to RL Reasoning: A Budgeted Search Perspective [arXiv] [episode]
  45. Group Adaptive Clipping Policy Optimization [arXiv] [episode]
  46. HBQ: Hierarchical Scaling Block Quantization with Hardware-Efficiency-Aware Design for Accurate LLM Inference [arXiv] [episode]
  47. ReNFT: Repairing Mode Collapse in Reward Post-Training via Internal Probability-Mass Recalibration [arXiv] [episode]
  48. Lightweight Adaptation of EEG Foundation Models for Stroke Motor Imagery Decoding: Domain Shift and Subject-Level Robustness [arXiv] [episode]
  49. Discovery and Characterization of Three New High-amplitude delta Scuti Stars from TESS Observations [arXiv]
  50. Skill Following: Evaluating Actual Skill Use in Retrieval-Enabled LLM Agents [arXiv] [episode]
  51. Dense Process Supervision for Search Agents via Fact Utility Estimation [arXiv] [episode]
  52. Nonparametric inference for density-dependent McKean--Vlasov diffusions [arXiv] [episode]
  53. GazeRefine: Expert Gaze as a Test-Time Prompt for Training-Free Medical Image Segmentation [arXiv] [episode]
  54. Foundation models for electricity price forecasting and battery arbitrage: Can they replace market-specific forecasting models? [arXiv] [episode]
  55. Towards Agentic Cloud Engineering: Graph and Loop Engineering with a Zero-Trust Agent Harness [arXiv] [episode]
  56. Do Multimodal LLMs See Before They Read? Diagnosing Contextual Sycophancy [arXiv] [episode]
  57. Right Frame, Wrong Rule: Cultural Cues Expose the Financial Knowledge Gap They Were Meant to Close [arXiv] [episode]
  58. JENGA: Exploiting Counter-Based RowHammer Countermeasures to Break Real-Time Predictability [arXiv] [episode]
  59. Griotte: Verified Compartmentalisation via Capabilities [arXiv] [episode]
  60. The quiescent states of V745 Sco and V3890 Sgr: VLT/X-shooter and Swift/XRT+UVOT observations [arXiv] [episode]

Listen

Everything the estate published on 2026-09-02, in the player. Playing from this page is how the edition learns whether anyone did.