Daily bookmarks: relevant news, facts and discoveries.
Articles 1216
Engineering ·
Published 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.
Engineering ·
A September 28, 2026 post describes how SIGReg, used by LeJEPA and LeWM, approximates a sliced MMD whose expansion reveals repulsion between pairs of samples in a batch.
Engineering ·
Published on September 27, 2026, HySparse2 is described as a hybrid sparse-attention method with two-level KV sharing, aimed at efficient prefill and compact KV-cache storage for agents with long contexts and multiple interactions.
Engineering ·
An ICLR 2026 paper addresses efficient autoregressive inference for probabilistic Transformer models and, according to the post, a decoding bottleneck.
Engineering ·
Published on September 27, 2026, the item describes LENS, a reinforcement-learning framework that jointly optimizes reasoning and text-guided image segmentation.
Engineering ·
A September 27, 2026 publication reports preliminary ablations suggesting V=K+M is an effective architectural choice; the result still needs validation at larger scales.
Engineering ·
A report claims that FairPlay-protected iOS IPAs can be decrypted without a physical iPhone or jailbreak. The claim may interest people studying app protection, but the briefing does not verify it.
Engineering ·
A paper on intra-kernel GPU tracing through binary-level instruction splicing; its preview describes probes inserted into kernels to record execution data.
Engineering ·
An article presents NSA, a trainable sparse-attention mechanism that combines algorithmic choices with hardware-aligned optimizations for efficient long-context modeling.
Engineering ·
A more elaborate workflow does not always improve outcomes. Compare components by task quality, consumption and data exposure.
Engineering ·
Published on September 26, 2026, the article compares standard Transformers and Mixture-of-Experts, with and without looping. The authors report that Looped-MoE scales better than the standard baseline, while dense-looping models do not.
Engineering ·
An article published on September 26, 2026 follows a vLLM request from queue to decoding and explains per-iteration scheduling and KV-cache block allocation.
Engineering ·
Internal representations do not make information anonymous. Define data boundaries before connecting multiple models.
Engineering ·
Published on September 26, 2026, Unsloth's guide ranges from beginner to advanced and covers using GRPO to train reasoning models.
Engineering ·
Published on September 26, 2026, the note presents JAZ, a framework using a single recursive code primitive for agent loops, and reports a claim of lower cost and better performance than two alternatives.
Engineering ·
Fewer components can ease maintenance. Data and action boundaries must remain explicit at every step.
Engineering ·
Published on September 26, 2026, the work presents a theoretical criterion for feature-map spikiness and a channel-wise polarity-aware feature map. The authors evaluate the approach across six families of visual tasks.
Engineering ·
A study uses program-generated byte sequences and an adaptive generator to investigate whether systems can learn transferable structures without training on natural data.
Engineering ·
Published on September 26, 2026, the paper presents an LLM compression method using singular value decomposition (SVD) that updates compressed weights after truncation.
Engineering ·
An article reports three kinds of computational redundancy across loops in looped Transformers. The authors say exploiting them reduces latency by 1.65× and memory use by 6×.
Looking for something else?DocumentationContact
