Diffusion models amplify demographic biases present in training data, and current debiasing methods usually require costly attribute annotations, model retraining, or sacrifice output diversity. This paper proposes Debias Anything, a training-free guidance method that steers both unconditional and text-conditional diffusion models toward arbitrary target proportions for any sensitive attribute described in free text. By linking the frozen denoiser to a vision-language embedding space via a lightweight adapter, it jointly enforces batch-level fairness through optimal assignment and restores diversity through a semantic disagreement score, offering broad applicability without labeled data.

Long-term dependencies pose a fundamental challenge for sequential decision making because Transformers suffer quadratic attention scaling and RNNs lack expressive hidden states. This paper proposes Decision Titan (DTi), which grafts Titans-style Test-Time Training layers onto Decision Transformer to store episodic memories in fast weights updated by gradient descent at both train and test time. The authors introduce X-Maze, a controlled extension of T-Maze, and show that DTi can learn dependencies 20× longer than its context window and extrapolate to unseen episode lengths, while visualizing gate dynamics to explain how momentum propagates long-range information. Although the experiments are limited to synthetic memory tasks, the mechanistic ablations and honest characterization of catastrophic forgetting provide a valuable foundation for future research.

Contextual linear optimization faces a discontinuity: small prediction errors can flip the LP solution to a different vertex, making regret non-differentiable pointwise. This paper proves that, after averaging over the data distribution, the population regret is locally quadratic with curvature concentrated on the walls of the normal fan. The authors derive the exact curvature measure $K_0$, provide a tractable single-projection approximation $\mathsf{K}_{\lambda,s}$, and apply it to scenario generation, showing a $30.8\%$ regret reduction on a battery arbitrage task.

Medical imaging models often fail when deployed across scanners or hospitals because they overfit to domain-specific textures and appearance statistics. PhaseAT addresses this by adversarially perturbing the Fourier phase of the luminance channel while keeping the amplitude spectrum fixed, generating hard training views that stress spatial organization under matched style. By encouraging the model to ignore brittle amplitude-driven correlations and focus on stable geometric structure, PhaseAT achieves large gains on both single-source and multi-source domain generalization benchmarks.

Can response safety be scored by cosine similarity to a safe-response embedding centroid? The paper shows that such a raw positive prototype identifies a location, not a separating direction, and formalizes why positive-only observations cannot determine the sign of the safe-unsafe gap. On three prompt-controlled safety corpora and four frozen encoders, the unreferenced prototype is weak or inverted while an explicit safe-minus-unsafe reference recovers useful ranking; the result is carefully scoped to the centroid rule, not all one-class methods.

CI pipelines for LLM training frameworks are drowning in GPU-heavy end-to-end tests that can take hours per run. FastCI attacks this bottleneck by replacing coarse file-level mappings with a runtime evidence graph built from lightweight stack sampling, then pruning redundant tests via caller/callee context, scheduling risky tests first, and shrinking workloads along dimensions outside each test's validation scope. Deployed at ByteDance, it claims a 77.5% latency reduction and 63.9% GPU savings, though the evaluation rests on an internal codebase with no open-source artifacts.

SkillEvoLean tackles the challenge of improving LLM agents for Lean theorem proving without updating model parameters. The core idea is to evolve natural-language procedural skills through progressive refinement from execution trajectories and, crucially, a mutation operator that perturbs skills in a semantic concept space to escape local optima when all proof attempts fail. This matters because existing skill-evolution methods stagnate on hard theorems where no sampled trajectory succeeds, leaving the solver with no signal for improvement.

Does RL post-training create new agentic capabilities or just redistribute probability mass toward high-reward behaviors that a base model already possesses? This paper shows that, across 14 base/post-trained checkpoint pairs and three process-centric agentic benchmarks, pre-trained base models equipped with a lightweight harness often surpass their post-trained counterparts in solution coverage (pass@$K$) given enough test-time compute. To diagnose this coverage-accuracy trade-off, the authors propose Sharpening Tax, a single-number metric for the loss in test-time scalability after post-training, and present Posterior-Tempered Group Sampling (PTGS), a Bayesian sampler that mitigates the tax during RL training.

Emergency Department triage assigns acuity scores that can be skewed by non-clinical patient descriptors. This paper asks whether open-source LLMs change their pediatric Emergency Severity Index predictions when only demographic, socioeconomic, or contextual variables are injected into otherwise identical clinical vignettes. The authors audit ten models and find that a small fine-tuned Qwen2.5-7B is the most counterfactually stable, challenging the assumption that larger or medically pretrained models are necessarily fairer. The study matters because locally deployable open-source LLMs are increasingly considered for privacy-preserving clinical decision support, yet their fairness profiles across model families were poorly understood.

SIEVE tackles selective unlearning in vision-language models (VLMs), where a model must forget specific sensitive image-question-answer associations about an individual while retaining permitted facts about the same person. The core idea is to suppress attention-value representations for forget examples toward zero and anchor retain-example values to a frozen reference model, coupled with sequence-level negative cross-entropy. This matters because VLMs can recover biographical information from faces, creating privacy risks, and prior methods struggle to disentangle closely related forgotten and retained knowledge without degrading utility.

Guided diffusion forces a costly trade-off between enforcing requirements and preserving sample quality, usually resolved by small guidance scales and long sampling trajectories. This paper introduces LEEGS, which learns a time-dependent guidance schedule end-to-end by backpropagating the guidance loss through the sampling trajectory. A cheap gradient approximation—treating the guidance gradient as a stop-gradient—cuts training time by roughly 4$\times$, letting the method train on only 30 examples and match or beat strong baselines across image inverse problems, face-ID guidance, and PDEs with as few as 50–100 NFEs.

Quantum key distribution deployments provision security margins assuming stationary channels, yet physical devices drift between calibrations. This paper frames adaptive eavesdropping under trusted channel noise and device drift as a constrained Markov decision process, where an attacker selects compact learnt quantum circuits each round while the noise follows an Ornstein–Uhlenbeck process. By bounding the adaptive value between the best fixed circuit and a dynamic-programming upper bound, the authors let operators measure how conservative their margins are and tune recalibration intervals, addressing a key economic barrier to QKD adoption.

Molecular tumor boards must distinguish truly unsupported therapies from evidence-supported recommendations that still require oncologist review because of incomplete clinical information or poor performance status. This paper introduces OpenMTB-Audit, a 500-case synthetic NSCLC benchmark with four safety labels, and exposes pervasive over-refusal: every tested LLM configuration collapses the Partially Supported label, misclassifying valid-but-caveated therapies as either Unsupported or Supported. To address this, the authors propose MTB-AuditAgent, a deterministic seven-module pipeline that reduces over-refusal to 6.7% and achieves 91.2% accuracy. The work matters because label collapse in high-stakes oncology auditing can conceal unsafe endorsements or inappropriately block appropriate treatments.

iADD tackles the alignment-diversity trade-off in reinforcement-learning finetuning of diffusion models such as Stable Diffusion. Its core idea is to update early denoising timesteps rather than late ones, combined with a sparse incremental curriculum and Feynman–Kac particle branching during training. Better alignment without mode collapse matters because existing reward-optimization methods frequently degrade visual quality and diversity.

The paper isolates matrix-valued transport from the broader sheaf-diffusion framework by casting it through quiver representations. Its core insight is that standard multi-head attention implements diagonal transport over the head space, and that allowing edge-dependent off-diagonal routing before aggregation yields a strictly richer message-passing primitive. The authors introduce Topological Attention (Top-A), which adds cross-head communication while exactly recovering vanilla attention when the learned routing vanishes. Benchmarks on relational, heterogeneous, and algorithmic tasks show that the gains are task-dependent: largest when individual interactions require distinct transformations, but not systematically present under heterophily alone.

Multi-agent motion planning with diffusion models currently demands costly training on large feasible-trajectory datasets, making learned planners map-specific and data-hungry. This paper introduces TFDP, a training-free diffusion-style planner that replaces global neural score networks with analytical local scores derived from smoothness, obstacle-avoidance, inter-agent separation, and velocity-limit terms. By exploiting the spatial and temporal locality of trajectory refinement, the method scales to hundreds of agents and obstacles in seconds without any demonstrations or neural network training.

Comparing high-dimensional discrete distributions is notoriously difficult because the state space grows exponentially and differences can hide in complex joint dependencies. This paper introduces Zero Flux, a flow-based discrepancy that extends the continuous Zero Flow criterion to discrete state spaces. The core idea is to construct a discrete bridge between two distributions via coordinate-wise independent coupling, measure local probability fluxes at the midpoint, and show that all fluxes vanish if and only if the distributions are identical. This yields a decomposable, sample-based discrepancy with finite-sample guarantees, offering both a global comparison score and interpretable coordinate-wise contributions.

Personalization encoders compress lengthy user interaction histories into a fixed preference state for downstream ranking or generation, but this compression can suppress evidence that remains available in cached per-timestep representations. This paper proposes REPAIR, a lightweight module that identifies and aggregates corrective signals from a frozen encoder's own cache across long, short, and episodic temporal views, then adds the correction back to the compressed state before the task head. By keeping the host encoder and task head frozen, REPAIR offers a practical way to improve deployed personalization systems without retraining them.

TACO attacks the memory bottleneck of full-parameter LLM fine-tuning by replacing dense AdamW states with a geometry-induced sparse update. The core idea is to use a dimension-normalized $\ell_1 \to \ell_1$ operator norm, whose exact steepest-descent direction is a ternary, column-wise one-sparse matrix. By retaining only a tiny FP8 heavy-hitter history per column, TACO slashes persistent optimizer state by over $170\times$ and peak memory by nearly $3\times$, enabling 30–32B parameter models to fine-tune on a single 80 GB GPU.

IQS-BO tackles the per-iteration cost of Bayesian optimisation by replacing surrogate fitting and acquisition-function maximisation with a single transformer forward pass. The core idea is to cast query selection as supervised classification: given a finite candidate set, the model predicts which candidate maximises the objective, trained with cross-entropy on synthetic priors where the true best candidate is known for free. The method can run fully in-context or on top of a fixed surrogate, and it introduces a mixture prior of GP and non-stationary functions to improve optimisation performance.