Hashimoto and Kawachi study Hamiltonian function learning: predicting $\operatorname{Tr}[f(H)\rho]$ from classical descriptions of a Hamiltonian $H$ and state $\rho$ when the function $f$ is fixed and known. Their core result is a rigorous average-case classical hardness proof for two explicit instances—$f_{\cos,\pi}(\lambda)=\cos(\pi\lambda)$ and $f_{\exp,\beta}(\lambda)=e^{-\beta\lambda}$—obtained by reducing from the hardness of factoring random RSA moduli via Shor's algorithm embedded in projector Hamiltonians. This settles an open problem left by Morohoshi et al. and gives a computational quantum advantage in a quantum machine learning task that has already been demonstrated experimentally on superconducting devices.

GLoC-EHR addresses clinical outcome prediction and reasoning over structured EHRs by equipping a language model with two complementary memories: a compact global trajectory summary and a bounded local set of selected events. The architecture aligns both routes with clinical text through hospital-course generation and masked-concept description recovery, then learns record-grounded, evidence-cited reasoning via GRPO with rewards for answer correctness and lexical evidence matching. Its core contribution is showing that a small multimodal LLM can match strong direct-prediction AUROCs while producing interpretable rationales tied to the patient record.

This paper tackles the rigidity of standard Difference-in-Means (DiM) activation steering, which applies a single global vector to all activations and ignores nonlinear manifold structure. The authors propose Kernelized Activation Steering (KAS), lifting steering into a reproducing kernel Hilbert space so that each activation receives a locally adaptive, implicit update based on kernel similarities to source and target reference sets. The framework strictly generalizes DiM—recovering it under a linear kernel—and connects activation steering to maximum mean discrepancy witness functions, with empirical evaluations on jailbreaking, behavioral control, and image style generation.

Most fairness tests ask whether a model treats groups exactly equally, but real audits often need to decide whether a disparity exceeds a legally or practically meaningful tolerance. This paper formalizes tolerance-based fairness auditing as two distinct statistical objectives—formal violation certification and early-warning sensitivity screening—and develops empirical-likelihood procedures calibrated to each. Constrained empirical likelihood (CEL) controls false violation declarations over a tolerance region, while split empirical likelihood (SEL) and an adjusted variant (ASEL) use an adaptive boundary-proxy principle to increase detection near the boundary. The work matters because it translates abstract fairness requirements into actionable, objective-dependent statistical guarantees.

Treating model release as a design problem rather than a convention, this paper asks how many of $k$ automated tests a model should pass before deployment to guarantee a stated reliability target while retaining as many good models as possible. Using a two-class Gaussian latent-factor model with common within-class correlation $\rho$, the author proves that precision rises strictly with the required pass count $r$, so the optimal gate is the most lenient $r$-of-$k$ rule that meets the target. This matters because teams often default to pass-all gating, which can decimate the pool of acceptable models when tests are even modestly correlated; the paper shows that misjudging dependence by a factor of two can hurt more than shifting the gate by one test.

The paper identifies a failure mode in the standard RLVR pipeline: training one policy with reinforcement learning from verifiable rewards (RLVR) improves single-sample accuracy but can hurt majority-vote performance because RLVR increases correlation among voter errors and collapses output diversity. The proposed fix is an "adapter thicket" — splitting the same training budget across $K$ independent LoRA adapters, each trained on a disjoint random data shard. Across 16 (model, $K$) settings and four models (1.5B–8B), thickets consistently out-vote the fully trained single adapter, and for $K \geq 4$ they protect the base model's vote while keeping most of the single-sample gain.

Detecting cryptic ligand-binding pockets from unbound protein structures is difficult because the pockets are often absent or closed in apo conformations. The authors propose an algebraic counting field (ACF) that maps explicit local structural features into compact, integer-weighted lookup tables so that every residue score can be mechanically reconstructed from training counts and spatial aggregation. The goal is to achieve competitive cryptic-pocket prediction while keeping the inference fully auditable, without relying on sequence homology, structural templates, or deep learning at inference time.

In model-predictive control with the cross-entropy method, a learned world model scores candidate action sequences and refits the Gaussian proposal for the next search iteration, creating planner-induced query shift. The authors disentangle these two roles by saving every CEM candidate pool from four surrogate models, executing all candidates in a resettable environment for a ground-truth ranking, and rescoring each pool with every model in a crossed design; they then splice the first elite set with environment-ranked elites to trace causal consequences through later iterations. Their results show that standard proposal distances contract even as ranking agreement stays near chance and mean disagreement grows, and that a single environment-guided elite update lowers final selected cost—demonstrating that world-model quality must be judged by its effect on the proposal, not just on candidate scoring.

This work investigates how Kolmogorov-Arnold Networks (KANs) scale with training dataset size across three recent variants—BSRBF-KAN, Gottlieb-KAN, and Faster-KAN—on MNIST, Fashion-MNIST, and a magnetic-parameter regression task. The authors fit test loss versus dataset size with a smoothly broken power law, report regime crossings from random-guess to faster- and slower-scaling branches in classification, and analyze how the shapes of learnable edge functions evolve as data grows. The central takeaway is that KAN scaling is architecture-dependent and task-dependent, and that Gottlieb-KAN edge functions collapse into a single shape family rather than diversifying symbolically.

HADRec targets two key gaps in AI-driven medication recommendation: drugs are usually treated as anonymous tokens, and predictions ignore the hierarchical structure of the ATC taxonomy. The proposed framework fuses LLaMA-7B–encoded clinical text with ChemBERTa molecular embeddings via cross-attention, then generates prescriptions through a four-level top-down ATC predictor constrained by a novel consistency loss. This is important because grounding recommendations in both chemical structure and medical ontology could improve accuracy, interpretability, and clinical trustworthiness.

This paper tackles endogenous nonstationary stochastic bandits where actions drive an unobserved latent linear system and rewards are bilinear in the action and latent state. Prior work achieved only $\widetilde{\mathcal{O}}(T^{2/3})$ regret via explore-then-commit (ETC) because the infinite-memory reward process couples learning and planning. The authors show that under stable dynamics the memory can be truncated geometrically, reduce the problem to a block-level linear bandit via cyclic approximation, and propose an optimistic block-UCB algorithm (Block-OFUL) that achieves $\widetilde{\mathcal{O}}(\sqrt{T})$ regret. The result is theoretically appealing but leans heavily on an exact block-maximization oracle and on knowing the dynamical stability constant $\rho$.

Long-horizon probabilistic forecasting of dynamical systems is computationally demanding because traditional flow matching requires integrating along an auxiliary generative time axis. This paper proposes Variational Streaming Flow (VSF), which augments a physical-time velocity field with a single latent variable sampled once per rollout, enabling efficient multimodal forecasting with one network evaluation per step. VSF improves long-horizon accuracy and distributional fidelity on chaotic and stochastic systems and can plug into existing JEPA world models to boost planning success.

RPTune addresses product search for small merchants whose catalogs fit inside long-context LLMs. Rather than relying on retrieval pipelines, the authors show that feeding the full catalog works only when items are learned to be pruned, ranked, and positioned with downstream LLM feedback, after which the LLM itself is post-trained on these curated contexts using purely synthetic supervision. The result is a fully automated pipeline that achieves large accuracy gains and latency reductions across 7 merchants and 9 model families.

STEER attacks the high inference cost of relational foundation models by shrinking their graph-sampled context using schema semantics instead of random dropping. The approach prompts an LLM once per task to rank foreign-key edges into relevance tiers, then probabilistically steers neighborhood traversal during sampling. Evaluated on RelBench across three state-of-the-art RFMs, it cuts average context size by about 40% while maintaining or slightly improving accuracy, offering a practical, model-agnostic optimization.

Test-time reinforcement learning (TTRL) lets language models adapt on unlabeled problems using their own samples as supervision, but a diffuse initial policy wastes the limited adaptation budget concentrating probability mass instead of converting existing capability into reliable output. This paper proposes entry-state sharpening: use data-free training before TTRL to place the checkpoint in a concentrated state that subsequent label-free adaptation can exploit more efficiently. The core insight is that checkpoint preparation should be treated as a state-control problem—where entry entropy is a label-free readout and reachable capability is the constraint—rather than as a choice of upstream training recipe.

Biological reasoning models fuse large language models with foundation-model representations and biological text to predict properties of DNA, proteins, and cells. This paper asks whether high benchmark accuracy actually reflects use of these biological inputs, or merely exploits overlapping text annotations. Through perturbation tests, evidence conflicts, linear probes, and reasoning-trace analysis across six models, the authors find that predictive representations are often ignored and that post-training gains can mask persistent over-reliance on text. The findings argue that evaluating input-specific reasoning must become standard alongside predictive accuracy.

Continuous diffusion language models have historically lacked two key features for practical deployment—variable-length generation and KV caching—while also trailing autoregressive and discrete diffusion in quality. This paper introduces Clock Diffusion, a framework that reparameterizes global diffusion time into position-dependent local clocks to enable semi-autoregressive denoising over token embeddings. Together with vectorized training masks and Cache Grab samplers, ClockDLMs close much of the perplexity gap on OpenWebText and unlock efficient inference, though downstream gains remain uneven across tasks.

In-context learning (ICL) lets LLMs adapt to new tasks from demonstrations, but reprocessing all examples for every query is costly and prior compression methods fail on complex, input-dependent tasks. This paper introduces Task Operator (TO), a training-free method that extracts per-head affine transformations — scaling weights and bias vectors — from the ICL forward pass and replays them during zero-shot inference by updating attention output projections. By treating the ICL effect as a dynamic, state-dependent modification rather than a fixed activation vector, TO substantially narrows the gap between zero-shot and few-shot performance while revealing sparse, task-specific circuits.

This paper tackles the problem that LLM-generated economic policy can sound plausible yet be dynamically inconsistent. It grounds an instruction-tuned language model inside six DSGE simulators via a common RL interface, using PPO and GRPO to learn multi-turn policy actions with delayed rewards. The core idea is to replace linguistic plausibility with simulated economic consequences, testing credit assignment, semantic-shift conditioning, and cross-simulator transfer. Results matter for any setting where language agents must make sequential decisions with long-horizon structural feedback.

Decision models like Jev return probability distributions to drive routing, abstention, and human review, yet average calibration cannot distinguish whether uncertainty stems from chance or from missing knowledge. The authors audit this distinction in a black-box decision model through roughly 575,000 API calls across paired interventions that vary supplied information while holding items fixed, plus targeted follow-up questions about the case and the model’s own familiarity. Their core finding is that confidence tracks correctness on familiar material but fails at knowledge boundaries—assigning up to $p_{\max}=0.80$ to salient options with no basis and growing overconfident past an observed temporal boundary—while apparent self-knowledge largely evaporates once surface cues are controlled. The work matters because it provides a practical, behavioral checklist for black-box knowledge audits and shows that querying case properties directly can be more reliable than reading a single output distribution.