SUPER INTELLIGENCE BREAK
A focused stream of intelligence from the sources you trust.
arXiv:2610.04378v2 Announce Type: replace-cross Abstract: Large language models increasingly mediate tool use in Model Context Protocol (MCP) systems, where adversarial influence may enter through user instructions, tool schemas, tool outputs, or protocol messages. Existing…
arXiv:2610.26834v1 Announce Type: cross Abstract: We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with a memory-revision…
arXiv:2610.26124v1 Announce Type: cross Abstract: Multi-teacher on-policy distillation (MOPD) aggregates the capabilities of multiple domain-expert teachers into a single student model. However, all domains in MOPD share the student's finite budget of optimization steps…
arXiv:2610.13073v2 Announce Type: replace Abstract: Processing a long sequence does not guarantee that distant information remains usable. As context grows, a fixed-size memory must retain many associations, connect information across positions, and preserve relevant…
arXiv:2610.23936v1 Announce Type: new Abstract: In-context learning (ICL) enables language models to make accurate predictions from demonstrations, but prediction success alone does not reveal the extent to which the underlying concept has been recovered.…
arXiv:2606.12629v4 Announce Type: replace Abstract: Because of polysemanticity and superposition, it has been challenging to attribute token understanding and generation to specific neurons or dimensions. Yet the information in the hidden states is so…
arXiv:2610.23935v1 Announce Type: new Abstract: Speculative decoding accelerates large language model inference through parallel verification of tokens proposed by a drafter. Diffusion drafters further parallelize proposal generation, but greater proposal capacity or higher mean…
arXiv:2512.01384v5 Announce Type: replace Abstract: Conformal regression provides finite-sample marginal coverage, but interval efficiency depends on how uncertainty is allocated across inputs. We propose emph{Conformal Laplace-Aware Predictive Scaling} (CLAPS), which combines a learned heteroscedastic…
arXiv:2610.23934v1 Announce Type: new Abstract: When using a large language model (LLM) to simulate a user population, how should we combine individual behavioral histories to predict the fraction of users who take a particular…
arXiv:2302.10578v2 Announce Type: replace Abstract: In fields such as medicine and drug discovery, the ultimate goal of a classification is not to guess a class, but to choose the optimal course of action among…
arXiv:2610.23862v1 Announce Type: new Abstract: We introduce a predictive world model for electronic health records (EHR) that represents patient histories as action conditioned latent trajectories. Rather than modeling clinical events solely through autoregressive token…
arXiv:2610.23864v1 Announce Type: new Abstract: Adaptive VLM ensembles save computation by stopping early on easy images and escalating harder ones. Although a single conformal threshold can make the final predictor marginally valid, it can…