The Morning Preprint

Tuesday 18 August

Showing the most recent batch — Tuesday 18 August. No papers have been ingested today yet.

432 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: 432 papers against a busiest day of 7974 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.

130Wed
7974Thu
406Fri
475Mon
432Tue

Show the answer

432 papers were ingested on Tuesday 18 August.

432 papers, three ways

Where today's batch landed

By the paper's primary arXiv category.
cs.LG106
cs.CL62
cs.CV35
cs.AI24
astro-ph.GA20
cs.CR15
stat.ML15
astro-ph.CO10

The index

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

  1. Subliminal Steering: Stronger Encoding of Hidden Signals [arXiv]
  2. The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures [arXiv] [episode]
  3. Wavelength Requirements for Life Detection via Reflected Light Spectroscopy of Rocky Exoplanets [arXiv]
  4. One Frozen Simulator Is Not Enough: Simulator Collapse in Multi-Agent RL [arXiv] [episode]
  5. Adam-HNAG: A Convergent Reformulation of Adam with Accelerated Rate [arXiv] [episode]
  6. WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantization [arXiv]
  7. RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies [arXiv] [episode]
  8. Evolutionary Ensemble of Agents [arXiv] [episode]
  9. LLMs as a Jury: Cross-Model Consensus Can Outperform Process Reward Models for LLM Reasoning [arXiv] [episode]
  10. Decentralized Federated Learning by Partial Message Exchange [arXiv] [episode]
  11. A Unified Bogoliubov Approach to Primordial Gravitational Waves: From Inflation to Reheating [arXiv] [episode]
  12. Near-Optimal Sample Complexity Bounds for Constrained Average-Reward MDPs [arXiv] [episode]
  13. Self-Supervised Learning of Graph Representations for Network Intrusion Detection [arXiv] [episode]
  14. Near-Equilibrium Propagation training in nonlinear wave systems [arXiv] [episode]
  15. Adaptive Optimization via Momentum on Variance-Normalized Gradients [arXiv] [episode]
  16. The Economics of AI Inference: Inflation Dynamics, Welfare Costs, and Optimal Monetary Policy under the Inference-Cost Phillips Curve [arXiv] [episode]
  17. The Energy Blind Spot: NVIDIA's Flagship Edge AI Hardware Cannot Support Process-Level Energy Attribution [arXiv] [episode]
  18. LoREnc: Low-Rank Encryption for Securing Foundation Models and LoRA Adapters [arXiv]
  19. Mitigating Bias in Locally Constrained Decoding via Tractable Proposals [arXiv]
  20. Infrared period-luminosity relations of Galactic Miras based on multi-epoch photometry and the Gaia parallax uncertainty [arXiv]
  21. Neutrino Flavor Conversion Shapes the Rate of Failed Core-collapse Supernovae [arXiv] [episode]
  22. Impact of subhalo dynamical friction heating on the formation of the first structures in the universe [arXiv] [episode]
  23. Non-KKT Accumulation in Entropic Mirror Descent [arXiv] [episode]
  24. HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam [arXiv] [episode]
  25. Deep Learning for BioImaging: What Are We Learning? [arXiv] [episode]
  26. Toward Hardware-Agnostic Quadrupedal World Models via Morphology Conditioning [arXiv] [episode]
  27. Interpreting "Interpretability" and Explaining "Explainability" in Machine Learning in Physics [arXiv] [episode]
  28. Knowing When to Defer: Selective Prediction for Responsible Knowledge Tracing [arXiv] [episode]
  29. Deep radiative zones affect the planetary cooling and internal structure: implications for exoplanet characterisation [arXiv]
  30. Constraint-Aware Generative Auto-bidding via Pareto-Prioritized Regret Optimization [arXiv] [episode]
  31. Scaling Machine Learning Interatomic Potentials with Mixtures of Experts [arXiv] [episode]
  32. MiDAS: A Multimodal Data Acquisition System and Dataset for Robot-Assisted Minimally Invasive Surgery [arXiv] [episode]
  33. Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents [arXiv] [episode]
  34. Learning symplectic model reduction based on a approximation theorem of symplectic embeddings [arXiv] [episode]
  35. SL-S4Wave: Self-Supervised Learning of Physiological Waveforms with Structured State Space Models [arXiv] [episode]
  36. Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning [arXiv] [episode]
  37. Technological Advances in Detecting and Managing Cognitive Impairment in Older Adults: Trends, Challenges, and Future Directions [arXiv] [episode]
  38. Directional Curvature from Armijo Backtracking: A Low-Cost Sharpness Probe and a Calibration-Free Learning-Rate Safeguard for Adam [arXiv] [episode]
  39. A Temporal Planning Framework for Disruption Aware Dynamic Route Optimization in Heterogeneous Railway Systems [arXiv] [episode]
  40. Shorter, but Still Trustworthy? An Empirical Study of Chain-of-Thought Compression [arXiv] [episode]
  41. MoRFI: Monotonic Sparse Autoencoder Feature Identification [arXiv] [episode]
  42. RETO: A Rotary-Enhanced Transformer Operator for High-Fidelity Prediction of Automotive Aerodynamics [arXiv] [episode]
  43. BiAxisAudit: A Novel Framework to Evaluate LLM Bias Across Prompt Sensitivity and Response-Layer Divergence [arXiv]
  44. RT-Lynx: Putting the GEMM Sparsity In a Right Way for Diffusion Models [arXiv] [episode]
  45. Red-Teaming the Agentic Red-Team [arXiv] [episode]
  46. Replacing Tunable Parameters in Weather and Climate Models with State-Dependent Functions using Reinforcement Learning [arXiv]
  47. Provably Efficient Personalized Multi-Objective Bandits with Proactive Conversational Queries [arXiv] [episode]
  48. Multilook Coherent Imaging: Theoretical Guarantees and Algorithms [arXiv] [episode]
  49. Gravitational Waves from Post-Inflationary Magnetism: Direct and Scalar-Induced Contributions [arXiv] [episode]
  50. Structural Generalization on SLOG without Hand-Written Rules [arXiv] [episode]
  51. CodeQuant: Unified Clustering and Quantization for Enhanced Outlier Smoothing in Low-Precision Mixture-of-Experts [arXiv] [episode]
  52. The Granularity Gap: A Multi-Dimensional Cross-Generational Audit of Sycophancy in Gemini Models [arXiv] [episode]
  53. In-Context Source and Channel Coding [arXiv] [episode]
  54. Closing the Loop: A Control-Theoretic Framework for Provably Stable Time Series Forecasting with LLMs [arXiv] [episode]
  55. Time-Correlated Video Bridge Matching [arXiv] [episode]
  56. Periodic Topological Deep Learning for Polymer Design and Discovery [arXiv] [episode]
  57. A Survey on Generative Model Unlearning: Fundamentals, Taxonomy, Evaluation, and Future Direction [arXiv] [episode]
  58. Misconception Diagnosis From Student-Tutor Dialogue: Generate, Retrieve, Rerank [arXiv]
  59. SAPO: Step-Aligned Policy Optimization for Reasoning-Based Generative Recommendation [arXiv] [episode]
  60. A Temporal Reasoning Benchmarking Framework for LRMs via Difficulty-controlled and Dynamic Test Generation [arXiv] [episode]

Listen

Everything the estate published on 2026-08-18, in the player. Playing from this page is how the edition learns whether anyone did.