Papers.Data.Code

Papers.Data.Code

The Weekly ML Signal

Selected, not collected.

9 issues · since August 2026

All issues

Weekly, monthly and quarterly digests of the top papers, repos and datasets.

2026

October 2026

  1. WeeklyOct 12 – Oct 18Issue #9Verifiable worlds and controlSynthetic control, executable world scoring, and theorem environments.
  2. WeeklyOct 05 – Oct 11Issue #8Long memory, adaptive computeStreaming video memory, token-wise decoding, and multimodal robot data.
  3. WeeklySep 28 – Oct 04Issue #7Planning, routing, and robust evalScore-first music generation, action scoring, and drift tracking.

September 2026

  1. QuarterlyQ3 2026Issue #6Latents, long context, open dataLatent decoding, long-context training, and auditable document corpora.
  2. WeeklySep 14 – Sep 20Issue #5Reasoning with executable tracesSearch-based math, tool use, and multi-turn environment data.
  3. WeeklySep 07 – Sep 13Issue #4Open systems, executable dataWorld models, exact decoding, and executable workflows with verified data.
  4. WeeklyAug 31 – Sep 06Issue #3Grounded evaluation gets harderLong-document QA, scientific reasoning data, and RAG scoring.

August 2026

  1. MonthlyAugust 2026Issue #2Auditable long-horizon systemsVerified terminal tasks, evidence-based evaluation, and action traces.
  2. WeeklyAug 17 – Aug 23Issue #1Verifiable agent environmentsExecutable worlds, skill-based evaluation, and reward-checked traces.

One useful ML email a week.

The Weekly ML Signal.

Papers.Data.Code

The Weekly ML Signal

Selected, not collected.

One useful ML email a week: the papers, repos and datasets worth your attention.

Worth your time, not all of your time.