Papers.Data.Code
The Weekly ML Signal
Selected, not collected.
Weekly edition · Oct 05 – Oct 11 · Issue #8
Long memory, adaptive compute
Streaming video memory, token-wise decoding, and multimodal robot data.
Top 3 papers
- Long-WAM: Scaling the Context of World-Action Models→
Unlike bidirectionally adapted WAMs, it preserves causal video pretraining, enabling 19.2-second histories to improve control.
- OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction→
It unifies proactive response timing and reusable caption memory in one causal decoder.
- ALoDLM: Adaptively Looped Diffusion Language Models→
It allocates denoising compute per token, letting an 8B DLM beat AR baselines while decoding 2.7× faster on GSM8K.
Top 3 repos
- nokia-applied-research/AnyJev→
Unlike recent typed-decision repos, it retrofits open LLMs into calibrated classifiers from one prefill, boosting BANKING77 auto-decidable coverage 7.7%→52.0% at ≤5% error.
- PSRben/VisionHOPE→
Unlike most vision backbones, it uses self-modifying memory updates and ships checkpoints across classification, detection, and segmentation.
- Tier-Flow/LoopVL→
Unlike most model repos, it releases reproducible recurrent VLM evaluation and visual-state analyses across 16 benchmarks.
Top 3 datasets
- ScholarCatalyst/ScholarCatalyst→
Unlike topical paper corpora, it benchmarks inspiration retrieval with 894 author-verified positives, hard negatives, and rationales.
- allenai/fetchman-data→
352,696 humanoid pickup episodes jointly align language, proprioception, actions, and head+wrist egocentric video.
- goodpjw2008/rsna-knee-2-5d-convnext-reader→
Releases end-to-end DICOM inference code and weights for a 0.929 public-LB knee MRI baseline.
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.