Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2609.40034v1 Announce Type: cross Abstract: Over the past decade, Machine Learning (ML) has been trained under dual objectives: minimizing prediction error via Empirical Risk Minimization (ERM)...
arXiv:2609.38543v1 Announce Type: new Abstract: Constantly evolving real-world knowledge necessitates models to be updated continuously. Especially in medicine, as clinical evidence changes over time...
arXiv:2609.38604v1 Announce Type: cross Abstract: Modern LLM agents increasingly tackle complex tasks through interactive, long-horizon exchanges with users, while existing benchmarks generally assum...
ReGain is a training‑free correction that improves subject fidelity in text‑to‑image diffusion models personalized with synthetic images. The authors show that fine‑tuning on synthetic images degrades fidelity due to inflated classifier‑free guidance, especially at high frequencies. ReGain measures this inflation per frequency band and scales it down during sampling, closing 51‑64% of the fidelity gap on Stable Diffusion v1.5 and improving performance on SDXL and SD 3.5 while preserving text alignment.
CellMSA introduces a novel single‑cell representation learning framework that leverages a multiple‑sequence‑alignment‑inspired context model. For each target cell, it retrieves relevant cells across batches and related cell types, summarizing cross‑cell patterns into a context‑dependent gene‑pair representation that is fed into a pair‑aware encoder. Pretraining on a massive human single‑cell corpus (≈109 million cells) and subsequent benchmarks demonstrate consistent performance gains over existing methods.
arXiv:2609.38925v1 Announce Type: new Abstract: Multimodal federated learning (MFL) has emerged as a pivotal paradigm for leveraging distributed data to enhance model performance. However, existing m...
arXiv:2609.38982v1 Announce Type: cross Abstract: Coding agents powered by large language models (LLMs) have shown remarkable abilities to autonomously reason about and achieve goals in the digital w...
arXiv:2609.39173v1 Announce Type: new Abstract: In this paper, we consider asymptotic properties of the support vector machine (SVM) in high-dimension, low-sample-size (HDLSS) settings under a spiked...
arXiv:2609.39229v1 Announce Type: cross Abstract: Automatic evaluation of faithfulness increasingly relies on a large language model acting as a judge, yet the most reliable judges are proprietary fr...
The paper investigates transferring specialized task-oriented behavior to a general language model without training or distillation. It applies two training‑free heterogeneous merging techniques—Intersection‑Merge (IM) and Activate‑Prune‑Merge (APM)—to project a specialist donor into the recipient’s parameter space and interpolate backbone weights. Experiments across embedding, reranking, reward modeling, and MoE code‑specialist tasks show that both methods improve the general model, demonstrating that simple parameter‑level merging can transfer capabilities across diverse specialist roles.
arXiv:2609. 39449v1 Announce Type: new Abstract: Distributionally robust optimization (DRO) studies parameter estimation under uncertainty in the underlying probability distribution and has emerged as a principled framework for analyzing robustness and generalization.
arXiv:2609.39684v1 Announce Type: new Abstract: Privacy has become an increasingly important concern in the Document Analysis community, to the extent that in many environments such as archives, gove...
arXiv:2609.39701v1 Announce Type: new Abstract: Value steering should change an LLM's normative priorities while preserving the scenario, facts, and task constraints underlying its answer. Convention...
arXiv:2609.39836v1 Announce Type: new Abstract: Contrastive vision-language models map visual and textual representations into a shared normalized embedding space, making cosine similarity the natura...
arXiv:2609.40035v1 Announce Type: new Abstract: The policy-gradient theorem gives the exact gradient under the current policy, but finite on-policy samples may miss rare high-return trajectories. We...
The study evaluates whether a fine‑tuned open‑weight model (Gemma‑3‑12B) can match the performance of GPT‑4o in extracting multi‑label intracranial hemorrhage acuity from non‑contrast head‑CT reports. Using a 2×2 design that varied adaptation strategy (classification head vs. instruction fine‑tuning) and training‑data source (distilled real GPT‑4o labels vs. synthetic GPT‑4o‑generated reports), the distilled instruction‑tuned model achieved macro‑F1 scores comparable to GPT‑4o and surpassed the untuned base model. The key finding is that the source of training data—distilled real reports—was more important than the fine‑tuning method, and that the entire fine‑tuning and inference process fits on a single 24 GB consumer GPU.
The paper presents efficient approximations for key statistics of the Neural Tangent Kernel (NTK) in finite-width neural networks using randomized trace estimation (Hutch++). It demonstrates that the NTK trace, Frobenius norm, effective rank, and alignment can be estimated with high accuracy via matrix-free products, leveraging the NTK’s positive-semidefinite structure to use one-sided estimators with forward or reverse-mode differentiation. Experiments on MLPs, GRUs, and a 410‑million‑parameter Transformer show orders‑of‑magnitude speedups and enable practical state‑space NTK diagnostics at large scales, including applications to RNN training and data‑scarce knowledge distillation.
arXiv:2605.15692v2 Announce Type: replace-cross Abstract: We study episodic reinforcement learning with fixed reward and transition functions, but with episode-dependent admissible action sets that a...
arXiv:2605.25663v2 Announce Type: replace-cross Abstract: Black-box adversarial attacks that minimize only the ground-truth confidence suffer from class drift: perturbations wander through the featur...
CollageAttack is a black‑box jailbreak that exploits cross‑modal alignment flaws in text‑to‑image models by shifting harmful semantics into the image plane. It combines context‑relevant scenes, scene‑grounded textual carriers, and spatially distributed text fragments to produce images that reveal hidden harmful meaning. Experiments on both open‑weight and commercial models show success rates up to 86.0%, outperforming the strongest baseline by 18.5 percentage points and consistently generating more harmful outputs while preserving the source intent.