Papers.Data.Code
The Weekly ML Signal
Selected, not collected.
Monthly edition · August 2026 · Issue #2
Auditable long-horizon systems
Verified terminal tasks, evidence-based evaluation, and action traces.
Top 3 papers
Tracked since posting
- Stealing Reasoning Traces from Proprietary LLM APIs→
Exploits weaker sibling models as cross-provider decryption oracles, recovering 182 credentials from 315,320 reasoning blobs.
- Recursive Synthesis for Long-Horizon Terminal Tasks→
Recursively grows sandbox-verified terminal tasks from seeds, yielding 37,484 tasks at $0.05 each.
- Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence→
Targets autonomous theory-level mechanistic discovery, grounding experiments in 13k interpretability papers and 32 tools.
- →
Top 3 repos
Tracked since posting
- ApodexAI/FrontierAgent→
Unlike most agent runtimes, it ships built-in evaluation and shows parallel Agent Team beating ReAct 38.5 vs 34.4 on APEX.
- MirroS-Lab/HarnessEval-W→
Unlike scalar-only video benchmarks, it uses model-agnostic skill routing with auditable evidence trees.
- TencentARC/SCoPE→
Adds explicit camera-ray token coordinates to a pretrained video DiT with under 0.1% new parameters.
- →
Top 3 datasets
Tracked since posting
- hamzabagirsakci/turkish-court-decisions→
It offers 11 million CC0 full-text Turkish court decisions spanning 1962–2026 across five court systems.
- Anthropic/claude-protein-binder-design→
It links 1,440 de novo binders to raw two-vendor kinetics, structures, and stepwise design provenance.
- PatronusAI/figmatrace→
Unlike final-design datasets, it captures 3,469 expert Figma action-frame trajectories over 126 long-horizon tasks.
- →
And this week
- PaperEnvHarness: Awakening Static Worlds for Agent Learning→
Unlike papers generating new environments, it retrofits existing verifiable ones via rollout-driven modular wrappers.
- RepoApodexAI/FrontierAgent→
Also #1 in this month's top repos
Picked this week and already among the month's top selections.
- Datasethamzabagirsakci/turkish-court-decisions→
Also #1 in this month's top datasets
Picked this week and already among the month's top selections.
One useful ML email a week.
The Weekly ML Signal.
Worth your time, not all of your time.
We track new ML work across papers, code, and data, then keep only the pieces worth a closer look. Selected, not collected.