Knowledge you can apply
Daily bookmarks: relevant news, facts and discoveries.
A September 29, 2026 entry describes KITE, a scaling paradigm combining scaling with upcycling and expanding decoder components that reuse a KV cache generated by a smaller prefiller.
Read articleLong context can extend exposure. Separate what an agent needs to remember from the identity of people mentioned in a conversation.
Read articlePublished on September 29, 2026, the MiMo RL Environment Explorer is described as an interactive space to inspect MiMo-V2.6 reinforcement-learning environments and run rollouts.
Read articleA publication dated September 29, 2026 describes a 7B model that continuously updated its weights using prediction-market data and ran continuous backtests over 390 days. It says the model outperformed several larger models in the same test environment.
Read articleThe approach distills linear attention models in long contexts and, according to the report, fully recovers full-attention retrieval while reaching 83–93% of mathematical reasoning performance with 3 billion tokens.
Read articleA paper published on September 29, 2026 tests on-policy distillation with one query across mathematics, coding, instruction following, and agentic tool use.
Read articlePublished on September 29, 2026, PTXBench evaluates model-generated matrix multiplication and attention kernels in architecture-specific PTX for H100 and B200 GPUs.
Read articleThe September 29, 2026 entry describes RESA, a residual-estimation proposal for recovering context missed when sparse attention selects only part of the KV cache. The post claims the technique adds no computational cost.
Read articleAnnounced on September 29, 2026, TIRx Harness is an open compiler harness for agents to explore, debug, and optimize GPU kernels, with compiler analysis, a catalog, hardware specifications, and remote evaluation.
Read articlePublished on September 29, 2026, the bookmark describes Vx's use of the `mlir!` macro to lower MLIR directly into leading-zero count instructions, avoiding manually maintained routines for different hardware.
Read articleAn isolated runtime controls what an agent can do. Privacy also requires controlling the content it sends to AI.
Read articlePublished on September 28, 2026, the post reports that kernel design agents generated and optimized Kimi Delta Attention kernels, reaching up to 2.96× the performance of FlashKDA on B300 with one-tenth the state error.
Read articlePublished on September 28, 2026, the article describes a hierarchical block-selection method for sparse attention in long contexts, reducing selection cost from O(N²) to O(N log N).
Read articleTurning speech into text changes the format, not the sensitivity. Evaluate every stage before introducing audio into a workflow.
Read articlePublished on September 28, 2026, the collection presents 200 environments for studying reinforcement learning in challenging, open-ended, long-horizon tasks.
Read articleA September 28, 2026 announcement presents open-source tools for pretraining, attention, checkpoint conversion, and language-model serving.
Read articleRunning a model close to data can change the architecture. It does not replace input processing, access control or real evaluation.
Read articlePublished on September 28, 2026, the post describes a technique that ranks neural representations or fine-tuning weight changes by their influence on behavior preserved under soft interventions.
Read articlePublished on September 28, 2026, the post presents Vx, which describes memory capacity, connections, and visibility to check programs. It reports two capacity errors found on a laptop without a GPU.
Read articlePublished on September 28, 2026, the material presents Self-Distilled Reasoner, an on-policy self-distillation approach for language models. Its preview describes student-sampled trajectories and dense teacher supervision at token level.
Read articleRota Nacional
30 days, no card, with a starting quota. After that, Pix credit from R$ 10,00.