Newshot
Papers 中文 Event #01M7S7WARD4AC1ADZF82SRYRJJ

JEPA as a Transferable Optimizer: Learning and Understanding Black-Box Optimization from Reward-Agnostic Pretraining

Event overview(model summary, not the source's own text)

JEPA as a Transferable Optimizer: Learning and Understanding Black-Box Optimization from Reward-Agnostic Pretraining — 3 reports from 2 sources so far.

How this story unfolded

Left: what it concluded — the key points the model distilled from the reports below, each one listing the reports it rests on. Right: what happened — ordered by time, each with a short summary so you can follow the thread without clicking through. The numbered links work both ways: tap a source under a point to jump to that report, tap the cited-by chip on a report to jump back to the point.

Key points0 items Model summary · ordered by when the evidence was published

  1. Only a handful of reports so far — the model has not distilled any key points yet.

ReportsBy publish time · newer further down

  1. November 19, 2026 · Thu
  2. Start

    JEPA as a Transferable Optimizer: Learning and Understanding Black-Box Optimization from Reward-Agnostic Pretraining

    Original summary arXiv:2611.23608v1 Announce Type: new Abstract: Black-box optimization evaluates candidates through simulations, rollouts, or experiments, often without gradients. Learned optimize

    arXiv cs.LG Original ↗
  3. Start

    Pragmatic Curiosity: A Unified Framework for Hybrid Learning and Optimization via Active Inference

    Original summary arXiv:2602.06104v3 Announce Type: replace-cross Abstract: Many engineering and scientific workflows rely on expensive black-box evaluations, requiring sequential decisions that mus

    arXiv stat.ML Original ↗
  4. November 20, 2026 · Fri
  5. +1d 0h

    Active Learning For Contextual Linear Optimization: A Margin-Based Approach

    Original summary arXiv:2305.06584v3 Announce Type: replace Abstract: We develop the first active learning method for contextual linear optimization. Specifically, we introduce a label acquisition a

    arXiv cs.LG Original ↗

Related reading

Counterfactual World Models: JEPAs That Reason About "What If"
arXiv:2611.24778v1 Announce Type: cross Abstract: World models operate in a world they cannot fully observe. When a hidd
arXiv stat.ML 11/19/2026 Full text here
77pts
JEPA4R: Embodied JEPA World Models with 4D Point Planning
arXiv:2611.19771v1 Announce Type: new Abstract: Joint-embedding predictive architecture (JEPA) world models predict futu
arXiv cs.RO 11/18/2026 Full text here
82pts
JEPA Dots: Dual-Latent World Model for Language-Guided Robot Control
arXiv:2611.24062v1 Announce Type: new Abstract: Predictive world models for language-guided robot control face a fundame
arXiv cs.RO 11/19/2026 Full text here
77pts