Papers.Data.Code
The Weekly ML Signal
Selected, not collected.
Weekly edition · Oct 12 – Oct 18 · Issue #9
Verifiable worlds and control
Synthetic control, executable world scoring, and theorem environments.
Top 3 papers
- S-Space: Exploring Spatial Workspace in Multimodal Language Models→
It identifies a fixed linear 3D spatial subspace in MLLM hidden states and causally edits it.
- DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models→
Prompt-aggregated reverse-rollout guidance plus trust-region restoration matches baselines using 12x fewer GPU hours.
- Player-1: Learning to Play from Synthetic Experience→
Uses synthetic RL across 16 verifiable games to learn transferable computer-use control, boosting MineDojo-300 success from 10.54% to 38.33%.
Top 3 repos
- NandhaKishorM/vegaml→
Unlike recent typed-decision repos, it adds local multimodal decisions with conformal sets and abstention.
- rockmatrix-ai/YANchor→
Unlike most long-context repos, it pairs a recurrent 4B model with O(1) memory and H100 serving that sustains 512 sequences at 12,431 tok/s.
- Optima-CityU/OpenArchEvo→
Searches executable PyTorch architecture programs beyond fixed NAS spaces, yielding spiking models with 31.7× lower estimated arithmetic energy.
Top 3 datasets
- 2077AIDataFoundation/TouchScale→
Unlike recent egocentric manipulation datasets, it adds synchronized 880-taxel full-hand tactile gloves across 100 released human-demonstration hours.
- HelenPeng/ExeWorldBench→
Targets executable Blender world construction, scoring functional correctness instead of visual similarity alone.
- FineEnvs/openai-math→
Provides 369 Lean 4 theorem-proving environments with exact-statement binary checking and hidden verified proofs.
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.