The Morning Preprint

Thursday 27 August

Showing the most recent batch — Thursday 27 August. No papers have been ingested today yet.

739 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.

199Fri
173Mon
431Tue
810Wed
739Thu

Show the answer

739 papers were ingested on Thursday 27 August.

739 papers, three ways

Where today's batch landed

By the paper's primary arXiv category.
cs.LG172
cs.CL123
cs.AI89
cs.CV49
cs.CR41
astro-ph.GA24
astro-ph.HE24
astro-ph.CO19

The index

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

  1. TokEval: A Tokenizer Evaluation Suite [arXiv] [episode]
  2. A Few Pages of Markdown: Committed AI Configuration and Lower Quality Cost after Coding-Agent Adoption [arXiv]
  3. From Memorization to Absorption: Mixed-Policy RL for Continual Knowledge Injection [arXiv]
  4. The "Curse of Knowledge" in LLM Query Simulation: Concept Provenance for Tracing Answer-Side Intrusion [arXiv]
  5. ShuttleArena: Interpretable Self-Play in Physics-Based Badminton [arXiv]
  6. What Do Medical Vision-Language Models Learn in Radiology? Transfer, Alignment, and Source-Proxy Leakage Under Distribution Shift [arXiv]
  7. What is the reason for the lack of cool evolved stars near the center of the Galaxy? [arXiv]
  8. Neural-Bayesian Structure Learning for Discrete Choice Modeling [arXiv]
  9. Hierarchical MoE for Multi-Modal ILD Diagnosis [arXiv]
  10. Mitigating LLM sycophancy with RL-based fine-tuning: Bayesian Truth Serum approach [arXiv]
  11. RWA-PoB: A Credential-Based Proof-of-Backing Framework for Tokenized U.S. Treasury Products [arXiv]
  12. A General Framework for Metropolis-Adjusted Dikin Walks: Dimension-Square Mixing on Polytopes and Log-Det Walks on Spectrahedra [arXiv]
  13. PhaseShift: Topology-Aware Data Harmonization and Model Consolidation Across Signalized Intersections [arXiv]
  14. Groundhog Bit-Flip Attack: Seeding Infinite Generation Loops in Mixture-of-Experts LLMs through Bit Flips [arXiv]
  15. Routed Graph Handoff: Adaptive Format Selection for Multi-Agent LLM Delegation [arXiv]
  16. The optimal redshift for dark energy II: application to cosmological data and the evidence for the phantom crossing of the CPL equation of state [arXiv]
  17. Provable Non-Acceleration of Standard Strang Splittings of Kinetic Langevin Dynamics [arXiv]
  18. Schr"odinger Bridges over Kinetic Swarming Models [arXiv]
  19. SHSP: Structure-Aware Hierarchical Solution Prediction for Mixed-Integer Linear Programming [arXiv]
  20. Generative Action-Chunk Sampling for Adaptive Stiffness Control in Physical Human-Robot Collaboration [arXiv]
  21. A Training-Free Proactive Defense Against Partial Speech Manipulation via Self-Embedding Steganography [arXiv]
  22. BixBench3: Benchmarking AI agents on research-study-scale computational biology tasks [arXiv]
  23. Combining Self-Embedding Audio Watermarking with Ultra-Low-Bitrate Neural Codecs [arXiv]
  24. InsightSR: Refining Symbolic Regression Search Spaces via Parallel Semantic and Structural LLM Guidance [arXiv]
  25. Centimeter-wave OH observations of comets 12P/Pons-Brooks and C/2023 A3 (Tsuchinshan-ATLAS) with FAST [arXiv]
  26. WAVE: Reversing the Guidance Hierarchy for Coarse-to-Fine Guided Depth Super-Resolution [arXiv]
  27. SAUSS: Stochastic Approximation with Unbiased Simulated Scores for Limited Dependent Variable Models [arXiv]
  28. Prefix-Denoising Consistency: Test-Time Verification for Diffusion Language Models [arXiv]
  29. Constraining gamma-ray burst viewing angles with Swift/XRT afterglow light curves [arXiv]
  30. Activation-Space Order-Swap Geometry: A Site-Asymmetry Audit [arXiv]
  31. LLMscope: Extracting LLM Assets from Edge AI Chips via Optical Probing [arXiv]
  32. FinRiskAtlas: Decision-Aligned Evaluation of Large Language Models for Financial Risk Review [arXiv]
  33. Two Dimensions Govern Agnostic Multiclass Transductive Learning [arXiv]
  34. Neither Precision Nor Architecture Alone: Controlled Tests of Failure Remedies for Physics-Informed Neural Networks [arXiv]
  35. Learning What to Share and What to Personalize: Hierarchical Strategy Co-Evolution for Agent Memory [arXiv]
  36. Provenance Before Prose: Claim-Locked Reporting for Statistical Text Generation [arXiv]
  37. Multi-messenger signal from QCD-Axion Ultracompact Minihalos [arXiv]
  38. A systematic study of AGN feedback in a disk galaxy using MACER. III. High Gas Fractions in AGN Hosts [arXiv]
  39. GUIDE: Generative Unsupervised Chinese Query Correction via Phonetic and Visual Shared-ID Encoding [arXiv]
  40. CoRE: Weakly Supervised Coarse-to-Fine Risk Evidence Learning in Driving Videos [arXiv]
  41. Short Horizons and Sparse Concepts: a Mathematical View of the Readout in the J-lens [arXiv]
  42. Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward Learning [arXiv]
  43. Laplace surface of eccentric orbits around Kerr black hole and black-hole effects on Lidov-Kozai cycles [arXiv]
  44. Escaping Low-Dimensional Overlap: Multi-Task Model Merging via High-Dimensional Sparse Disentanglement [arXiv]
  45. Right Values, Wrong Places: Diagnosing and Fixing Record Misplacement in LLM-Generated JSON [arXiv]
  46. Leveraging Speech Acts for Low-Data and Cross-Domain Conversation Derailment Forecasting [arXiv]
  47. Recovering the effective impact parameters in integral-field absorption-line tomography [arXiv]
  48. PaSta: Noisy Node Classification with Partial Label Learning [arXiv]
  49. CRAMER: Control via Request-Aware Masking for Editing Recommenders [arXiv]
  50. Capacity Overflow: A Blind Spot for Backdoor Attacks in Vision MoE [arXiv]
  51. GGSS: Geodesic-Gated Spherical Steering for Inference-Time Debiasing of Generative Vision-Language Models [arXiv]
  52. Adaptive Triggering for Bias Correction in LLM Reasoning [arXiv]
  53. Q&A or Document-Based? The Effects of Interface Type on How Screen Reader Users Access Interconnected Documents [arXiv]
  54. Large fraction of blends from Rowan et al. (2023) [arXiv]
  55. A meta-algorithm for ab initio reconstruction of complex mixtures in cryo-EM [arXiv]
  56. Refusal geometry reflects refusal training: diverse refusal prefixes can raise stable rank and weaken refusal vector ablation attacks [arXiv]
  57. OmniPhys: A Unified Multimodal Benchmark for Physics Understanding and Generation from Chinese Educational Corpora [arXiv]
  58. Can your AI agent be cheaper? Investigating the effects of task specifications on token spend in agentic coding tasks [arXiv]
  59. Broadband 12-26 GHz Radio Radiation Reveals Evidence for Micro-flares on AU Mic [arXiv]
  60. PIVOT: A Multi-Trajectory Dataset and Testbed for Pose, Intrinsics, and Novel Viewpoint Evaluation in Real-World 3D Reconstruction [arXiv]

Listen

Everything the estate published on 2026-08-27, in the player. Playing from this page is how the edition learns whether anyone did.