Tabular foundation models (TFMs) achieve strong zero-shot performance through in-context learning, but repeatedly encoding the labeled context at inference time is expensive. This paper studies how to distill TFMs into lightweight, dataset-specific students. It systematically examines two design choices—how to construct teacher supervision (full-context versus out-of-fold, mixed with labels or not) and whether cheap synthetic queries improve distillation—and derives a simple recipe that yields median inference speedups of $3.0\times$–$21.6\times$ while recovering roughly two-thirds of the teacher's accuracy advantage over a standard supervised student.

This paper tackles deterministic multi-impulse interplanetary rendezvous by embedding classical astrodynamics inside a reinforcement-learning policy. The proposed Reachability Analysis-Informed RL (RARL) framework maps bounded velocity perturbations into a local ellipsoidal waypoint set via the first-order position-from-velocity sensitivity matrix; the policy selects a waypoint, and a Lambert solver reconstructs the dynamically consistent impulsive arc. The core contribution is a physically structured action interface that supports both near-optimal nominal transfers and policy reuse across a family of dispersed departure states.

Bridging quantum-mechanical accuracy with mesoscopic length and time scales remains a grand challenge in materials modeling. This paper develops an ab initio phase-field framework that derives the mesoscopic evolution equation from molecular dynamics via Mori–Zwanzig projection, leaving only the free energy and mobility to be learned by neural networks from short machine-learning-interatomic-potential trajectories. The result is a data-efficient approach that propagates ab initio accuracy to scales far beyond direct atomistic simulation, demonstrated here on iron–boron melts and hydrogen–helium mixtures under planetary conditions.

Neuro-symbolic predictors require hand-crafted logical constraints, which limits adoption by practitioners without formal-methods expertise. This paper asks whether LLMs can translate natural language into propositional constraints that plug into NeSy predictors, and introduces auto-nesy-bench—an 18-task benchmark that measures both formalization fidelity and downstream accuracy. The core finding is that frontier LLMs can generate formulas approaching expert quality, and when the formulas are syntactically valid they often preserve downstream predictor performance, suggesting a plausible path toward making reliable NeSy AI more accessible.

Most LLM benchmarks treat generalization as aggregate accuracy on fixed prompts, which can hide per-example fragility when the same user intent is rephrased. This paper reframes generalization as behavioral stability across semantically equivalent input variations and proposes SAGO, a per-prompt, multi-axis framework that tracks changes in generation consistency, internal activations, confidence/uncertainty, and response mirroring. By evaluating eleven open- and closed-source models across six datasets and three variation families, the authors show that instability is heterogeneous, axis-specific, and not resolved by scale, challenging the assumption that high benchmark scores imply reliable generalization.

Standard gradient flow never reaches zero risk in finite time, while naive finite-time modifications drive the weights to the optimum with infinite speed. This paper formulates and solves a regularized least-time problem for the dissipation schedule $u(e)$ that controls the risk clock via $\dot{e}=-u(e)$. By minimizing $\int_0^{e_0}(\frac{\lambda}{2}|u'|^2+1/u)de$ subject to $u(0)=0$, the authors prove the optimal schedule is a cycloid and that it lands with the universal cusp $u^*(e)\sim c e^{2/3}$, placing the dynamics in the finite-time Hölder regime with vanishing weight speed.

Train–validation separation is the widening empirical gap between training and held-out accuracy during fine-tuning of pretrained models. This paper proposes a dynamic structural explanation: optimization progressively shifts from broadly reusable shared features that transfer well toward narrower, example-specific residual patterns that fit training data but generalize poorly. The authors formalize this intuition with a local mixture model and introduce fixed training-probe measurements of normalized absolute gradient-allocation dispersion, $D_{\mathrm{den},t}$, to observe this internal evolution without using validation data. Empirical tests span a controlled ResMLP hierarchy simulation, a 90-run NLP cohort across three backbones, and a 40-epoch ResNet-18 vision study.

Muon trains hidden matrices with a spectral-norm steepest-descent update but falls back to AdamW for embeddings, output heads, and auxiliary parameters. This paper focuses on the tied vocabulary table—one tensor aliased between sparse token lookup and dense softmax classification—and proposes AF-Muon, an AdamW-free Muon variant that assigns a finite-cap clipped-row linear-minimization oracle to the tied table, an RMS-normalized update to one-dimensional parameters, and the standard spectral update to hidden matrices. By eliminating all second-moment state, AF-Muon cuts optimizer-state memory by roughly 20% and improves mean validation loss across nine tied-token settings spanning text, image, and protein modeling.

CLEAR proposes a lightweight, task-agnostic post-hoc recalibrator for evidential deep learning that improves OOD and adversarial robustness without retraining the base predictor. Using held-out calibration data, it models group-conditioned latent geometry and measures conflict among perturbed latent views to detect unsupported evidence, selectively down-weighting overconfident predictions while preserving the original prediction. The method achieves strong gains across classification, regression, and object detection, though it trades off some in-distribution calibration and coverage for improved shift detection.

SupraTITO tackles the costly problem of predicting how peptide sequences drive supramolecular self-assembly. It learns transferable implicit transfer operators conditioned on sequence, topology, and periodic boundary geometry to generatively propagate many-molecule configurations over long physical lags. Evaluated on 400 coarse-grained dipeptide systems, it generalizes to held-out sequences and unseen concentrations while preserving molecular integrity through recursive rollouts. The work extends generative molecular dynamics from single-molecule conformations to collective periodic assembly.

This survey systematises the sprawling literature at the intersection of optimal transport and reinforcement learning through a four-question taxonomy: what role OT plays, which distributions are compared, which formulation is used, and how temporal structure is handled. By foregrounding the geometry-aware nature of transport-based discrepancies, the authors explain why OT can outperform $f$-divergences when learner and expert distributions overlap weakly—a common scenario in imitation learning, offline RL and robust control. The result is a broad methodological map spanning occupancy matching, trust regions, world-model learning and multi-agent settings.

The paper argues that attributing model behavior to synthetic training data requires first recording how each item was generated, because a waveform-label pair alone erases upstream history. The authors introduce a structured provenance substrate that binds scenario seeds, generated content, audio waveforms, fact checklists, quality signals, and review lineage into an immutable manifest. They instantiate this in a five-pass audit of 113 synthetic Japanese care-handoff speech assets totaling 1.552 hours, finding that while versioned, seed-disjoint manifests support some downstream attribution research, exact upstream tracing is blocked by floating generator aliases and missing per-item TTS stamps. The work reframes data attribution as a two-stage problem: establish a typed, versioned generation path $z \longrightarrow x \longrightarrow m \longrightarrow r \longrightarrow b$ before attempting causal or contributive claims.

This paper studies causal logistic bandits with counterfactual fairness constraints, where the learner observes only factual rewards and must infer counterfactual effects through shared logistic parameters. The core contribution is establishing minimax optimal regret-violation tradeoffs under a weak pooled coverage condition, showing that a target-specific information scale $V_\star$ governs the estimation difficulty and yielding matching $\widetilde{\Theta}(T^{2/3})$ upper and lower bounds.

Flow Matching (FM) suffers from trajectory crossings that create high-frequency velocity targets and amplify numerical integration errors during few-step sampling. This paper proposes CoFlow, a framework that injects a repulsive contrastive drift into the forward SDE to push training trajectories away from negative samples, thereby smoothing the learned velocity field. The authors derive a simulation-free stochastic interpolant, prove that the transformation preserves the endpoint distributions when boundary terms vanish, and show consistent improvements on ImageNet 256x256 at 20 sampling steps without extra training cost.

PPO is the workhorse of on-policy deep RL, yet it discards collected data after just a few updates, leaving substantial sample efficiency on the table. This paper asks whether reusing recent off-policy samples within a multiple importance weighting framework can improve PPO without changing its core mechanics. The authors propose two variants—uniform ($\omega$PPO-U) and balance-heuristic-corrected ($\omega$PPO-BH)—derive policy improvement bounds that account for stale advantages, and empirically show that moderate windowed reuse boosts final performance while excessive reuse can backfire.

User interactions across movies, news, and dialogue all supply evidence that can update a user model, even though their native vocabularies differ. This paper proposes an action-on-item schema that maps heterogeneous behaviors into a shared event interface, enabling one history encoder — PerTIDE — to learn reusable state updates for both predictive ranking and personalized headline generation. The authors support the idea with invariance and stability theory, plus controlled transfer experiments showing that a frozen source-trained core retains a $15.23$ MRR advantage over a random core, suggesting that compatible histories can share a single update mechanism.

This work tackles decentralized learning in multi-player stochastic games where each agent's state transitions depend only on its own state and action, but rewards are arbitrarily coupled. The core idea is to run independent online mirror descent in the dual space of local occupancy measures, using asynchronous dyadic confidence sets to handle unknown transition kernels without synchronized episodes. This matters because the guarantees scale with local state-action dimensions and mixing times rather than the joint product space, offering the first fully online, uncoordinated framework for this setting.

Diffusion language models generate text through iterative denoising rather than left-to-right prediction, making attribution a spatiotemporal problem: support for any output token is distributed across both network depth and timesteps. This paper introduces Diffusion Layer Integrated Gradients (DLIG), which extends Integrated Gradients to score a DLM's commitment to its completion at any chosen layer and denoising step, decomposing it into per-token contributions. It matters because existing token-attribution tools ignore the denoising trajectory entirely, while interventional alternatives are computationally expensive.

The paper tackles the problem of learning task-agnostic world models for planning in latent space. It introduces CTWM, a reconstruction-free method that combines residual prediction with a log-determinant regularizer to recover the scaled Laplacian representation, so that latent L2 distances correspond to environment commute-time distances. This matters because it enables zero-shot goal-reaching without reward, and empirically CTWM matches or beats the LeWM baseline while using half the parameters.

We study non-clairvoyant single-machine scheduling to minimize the sum of completion times, a problem for which no constant-competitive algorithm is possible in the pure setting. The authors introduce a robustness framework that leverages a classifier's confusion matrix to partition jobs into predicted classes, characterizing uncertainty as internal permutations rather than numerical scenarios. This structural information yields an oblivious non-adaptive Min-Cost algorithm that is simultaneously optimal for multiple criteria, alongside new adaptive and randomized extensions.