Interactive demo

Weekly curation

We do not pretend every paper is revolutionary. Here is a transparent slice of this week: top selections versus near-misses and why they did not make the shortlist.

Selected (sample)

Full week of 20 →

Near misses

  • 47

    Spatial Foundry: A Multimodal Dataset for Spatial Intelligence

    Below top cut for this week’s automated tranche

  • 47

    Auditing AI peer reviewers: a benchmark on real scientific papers

    Below top cut for this week’s automated tranche

  • 46

    Back to Pixels: Lifting Latent Video Diffusion into Pixel Space

    Below top cut for this week’s automated tranche

  • 46

    A deployed four-city entanglement-based QKD network with real-time key delivery, benchmarked against finite-key security bounds

    Below top cut for this week’s automated tranche

  • 45

    HEIR: Harness Egocentric Intent for Human-Robot Interactions

    Below top cut for this week’s automated tranche

  • 44

    Inner Product Regularization for Imitation Learning

    Below top cut for this week’s automated tranche

  • 42

    Informative Path Planning with Gaussian Processes: Error Bounds, Algorithms, and Benchmarks

    Below top cut for this week’s automated tranche

  • 41

    Internalizing Investment Mandates: A Neurosymbolic RL Approach to Portfolio Allocation

    Below top cut for this week’s automated tranche

  1. 1Discover — arXiv queries per hot topic
  2. 2Score — five-factor disruptiveness model
  3. 3Select — top ~20 + near-miss transparency
  4. 4Explain — fixed article skeleton for humans
  5. 5Serve — free HTML + x402 JSON for agents