Monday 17 August

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.

475 papers, three ways

Where today's batch landed

The index

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

1. Transformer Explainer: Learning LLM Transformers with Interactive Visual Explanation and Experimentation [arXiv]
2. Robust, randomized preconditioning for kernel ridge regression [arXiv] [episode]
3. Causal Falsification of Digital Twins [arXiv] [episode]
4. Optimal Transport for Machine Learners [arXiv] [episode]
5. Bayes-optimal learning of an extensive-width neural network from quadratically many samples [arXiv] [episode]
6. Resume Means Resume: A Machine-Checked Conformance Contract for Checkpoint, Interrupt, and Resume Semantics in Workflow Persistence Layers [arXiv] [episode]
7. High-dimensional Asymptotics of Denoising Autoencoders [arXiv] [episode]
8. A primer on optimal transport for causal inference with observational data [arXiv]
9. When LLM Agents Negotiate: Private Information and Dynamic Bargaining in Supply Chains [arXiv] [episode]
10. Universality laws for Gaussian mixtures in generalized linear models [arXiv] [episode]
11. Active Regression via Linear-Sample Sparsification [arXiv] [episode]
12. Proportional Committee Elections with Positive and Negative Votes [arXiv] [episode]
13. A variational Bayes approach to inference for low-dimensional parameters in high-dimensional linear regression [arXiv] [episode]
14. Understanding Alternating Minimization for Matrix Completion [arXiv] [episode]
15. CHIME: A Case for Efficient Long-Context Attention-FC Disaggregated Inference with DIMM-PIM [arXiv] [episode]
16. Discrete distributions are learnable from metastable samples [arXiv] [episode]
17. Topological complexity of spiked random polynomials and finite-rank spherical integrals [arXiv] [episode]
18. Langevin dynamics for high-dimensional optimization: the case of multi-spiked tensor PCA [arXiv] [episode]
19. Accelerated Markov Chain Monte Carlo Algorithms on Discrete States [arXiv] [episode]
20. Walrus: An Efficient Decentralized Storage Network [arXiv] [episode]
21. Quasi-Bayesian sequential deconvolution [arXiv] [episode]
22. TARL: Transaction-Aware Reliable Ledgers for Executable Memory Management in Long-Term Agents [arXiv] [episode]
23. UAV3DCrop: Benchmarking 3D Reconstruction in Repeated Multi-Angle UAV Crop Surveys [arXiv] [episode]
24. Distributional Autoencoders Know the Score [arXiv] [episode]
25. A phase transition between positional and semantic learning in a solvable model of dot-product attention [arXiv] [episode]
26. On the expressivity of deep Heaviside networks [arXiv] [episode]
27. SuperCoder: Assembly Program Superoptimization with Large Language Models [arXiv] [episode]
28. The Next Layer: Augmenting Foundation Models with Structure-Preserving and Attention-Guided Learning for Local Patches to Global Context Awareness in Computational Pathology [arXiv] [episode]
29. Learning k-body Hamiltonians via compressed sensing [arXiv] [episode]
30. Devlore: Device Interrupt Protection for Confidential VMs [arXiv] [episode]
31. AeroDPO: Unleashing Lightweight UAV Navigation with High-Fidelity Perception and Automated Preference Optimization [arXiv] [episode]
32. A New First-Order Meta-Learning Algorithm with Convergence Guarantees [arXiv] [episode]
33. MOSAIC: Adversarial Co-evolution of Specialist Heuristics and Problem Instances for LLM-based Automated Heuristic Design [arXiv] [episode]
34. Subspace clustering in high-dimensions: Phase transitions & Statistical-to-Computational gap [arXiv] [episode]
35. When Large Language Models are More PersuasiveThan Incentivized Humans, and Why [arXiv] [episode]
36. Humans are Missing from AI Coding Agent Research [arXiv] [episode]
37. NuOscProbExact: a general-purpose code to compute exact two-flavor and three-flavor neutrino oscillation probabilities [arXiv] [episode]
38. Subtype Robustness Is Not Just Accuracy: Calibration Under Unseen Subtype Shift [arXiv] [episode]
39. Insecurity Through Obscurity: Veiled Vulnerabilities in Closed-Source Contracts [arXiv] [episode]
40. NeuroPilot: An Agent-Driven Smart Pipeline for Processing, Quality Control, and Managing Neuroimages [arXiv] [episode]
41. Forecasting Side Effects of Activation Steering [arXiv]
42. The Judge Knows When It Knows: Calibrated Abstention for LLM-Based A/B-Test Prediction [arXiv] [episode]
43. Efficient transformer adaptation for analog in-memory computing via low-rank adapters [arXiv] [episode]
44. A Survey on Inference Optimization Techniques for Mixture of Experts Models [arXiv] [episode]
45. Generative Modeling with Bayesian Sample Inference [arXiv] [episode]
46. EchoChange: A Diffusion Language Model with Dual Pass Remasking for Factual Remote Sensing Disaster Change Captioning [arXiv] [episode]
47. Cueless EEG imagined speech for subject identification: dataset and benchmarks [arXiv] [episode]
48. Generative Brownian Bridge Diffusion In Motion Space For Enhanced Myocardial Strain Analysis [arXiv] [episode]
49. Beckmann Transport Models: From Autonomous Flows to One-Step Maps [arXiv] [episode]
50. ReXrank: A Public Leaderboard for AI-Powered Radiology Report Generation [arXiv] [episode]
51. Explainability in Practice: A Survey of Explainable NLP Across Various Domains [arXiv] [episode]
52. Protecting Creative Writing Copyright against AI Imitation via Implicit Watermarking [arXiv] [episode]
53. Enhancing Knowledge Tracing through Leakage-Free and Recency-Aware Embeddings [arXiv] [episode]
54. NL2SHACL-Bench: A Benchmark Suite for Natural Language to SHACL Translation [arXiv] [episode]
55. A Markovian Model for Learning-to-Optimize [arXiv] [episode]
56. Thermalizing Stochastic Programs [arXiv] [episode]
57. A dataset of high-resolution plantar pressures for gait analysis across varying footwear and walking speeds [arXiv] [episode]
58. Harnessing the Synergy between LLM Agents and Knowledge Graphs for Urban Socioeconomic Prediction [arXiv]
59. Large Language Models are Powerful Electronic Health Record Encoders [arXiv] [episode]
60. Primordial black hole induced gravitational waves in f(R) gravity [arXiv]