Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2608.29326v1 Announce Type: cross Abstract: Positive psychology dialogue aims to support emotional distress and positive resource building, requiring models to produce not only empathetic repli...
arXiv:2608.30585v1 Announce Type: new Abstract: Large language models are trained to follow instructions while refusing harmful requests. Jailbreaks exploit this balance to elicit content a model wou...
arXiv:2608.28843v1 Announce Type: cross Abstract: We show that smooth two-layer feed-forward networks (FFNs) expose an additional structural model extraction channel under a chosen-input raw-output o...
arXiv:2608.30719v1 Announce Type: new Abstract: Productive dialogue alignment requires distinguishing \emph{surface coordination} (acknowledgments and smooth task progression) from \emph{epistemic al...
arXiv:2608.30636v1 Announce Type: new Abstract: Optimizing absorption, distribution, metabolism, and excretion (ADME) is an important part of small molecule drug discovery. Many machine learning mode...
arXiv:2505.22255v4 Announce Type: replace Abstract: Sparse Autoencoders (SAEs) decompose language-model activations into sparse, interpretable features, but standard encoders usually treat the latent...
arXiv:2608.29510v1 Announce Type: cross Abstract: Aerial object detection is increasingly deployed in real-world applications, but models remain vulnerable to physical, universal adversarial patches...
arXiv:2608.30912v1 Announce Type: new Abstract: Artificial intelligence (AI) and natural language processing (NLP) are increasingly used to extract, integrate, and interpret biomedical knowledge rele...
arXiv:2608.30703v1 Announce Type: cross Abstract: Runtime guardrails are essential for reliable large language model (LLM) deployment, yet existing approaches typically rely on independent, external...
SpanCalib-VLM is a hybrid system for detecting hallucinated text spans in Vision‑Language Models. It combines a multimodal sequence tagger (XLM‑RoBERTa‑Large + SigLIP) with a fine‑tuned generative VLM (Qwen3.5‑4B‑SHROOM‑SFT) and uses a Union‑Calibrated Fusion strategy to re‑score candidate spans. On the SHROOM‑Visions English evaluation split, the ensemble achieves a Pearson calibration correlation of 0.41, an overall IoU of 0.39, a clean‑response IoU of 0.91, and a detection accuracy of 70.7%.
arXiv:2608.29001v1 Announce Type: new Abstract: In a data-driven world, efficiently organizing and mapping relationships between objects is crucial. Graphs are powerful tools for modeling these conne...
The paper introduces Stratified Consistency Distillation, a fine‑tuning method to improve natural‑language to logical‑formula translations for neurosymbolic reasoning. It generates multiple candidate translations with a large LLM, clusters them by semantic equivalence, and selects pseudo‑labels using entropy‑based strategies—majority voting for low entropy, LLM-as‑Judge for medium entropy, and unification or abstention for high entropy. The selected pseudo‑labels are then used to fine‑tune a smaller model, yielding significant gains in Pass@K and a new Equivalent Logical Similarity metric.
The paper introduces TACS, a trajectory‑aware candidate selection framework designed to improve jailbreak suffix optimization for large language models. Traditional gradient‑based methods choose candidates based solely on the lowest current loss, which the authors argue is myopic and can lead to reward hacking. TACS augments per‑step evaluation with a trajectory‑aware proxy, reference‑policy regularization, and a chi‑squared correction to encourage selections that remain effective beyond the immediate step. Experiments on HarmBench show that TACS consistently outperforms strong baselines, achieving higher attack success rates and more stable optimization behavior.
arXiv:2601.22118v3 Announce Type: replace Abstract: Artificial Intelligence (AI) has been on the rise in many domains, including numerous safety-critical applications. However, for complex systems in...
IndicDetect is a benchmark for evaluating AI‑generated text detection in Hindi, Telugu, and Tamil. It pairs curated human‑written texts with LLM‑generated counterparts across multiple domains and generators, testing detectors under domain shift, generator shift, and adversarial perturbation. The study shows that supervised neural detectors fail to generalize to unseen generators and attacks, with Hindi experiencing the greatest degradation, indicating that robustness—not peak accuracy—is the main weakness in Indic language detectors.
arXiv:2608.30646v1 Announce Type: cross Abstract: Reliable uncertainty estimation is a crucial requirement for deploying large language models (LLMs) and vision-language models (VLMs) in safety-criti...
A prominent paradigm in inference-time alignment employs lightweight supervisors to steer Large Language Models (LLMs). Through empirical analysis, we identify a structural mismatch in this paradigm:...
Machine unlearning for large language models (LLMs) often assumes that a pre-defined forget set matches what the model has memorized, but this frequently breaks in realistic privacy settings where the...
The next generation of mobile networks is envisioned as fully AI-native, with AI-RAN architectures embedding small language models (SLMs) to perform reasoning over real-time telemetry. The state-of-th...
Does unrestricted AI access bypass the cognitive effort required for learning, or does it streamline knowledge acquisition? This paper reports on a study where we compare three designs for user-AI int...