Research
Towards Mitigating Deceptive Safety Alignment in Large Reasoning Models (opens in a new tab)
arXiv cs.AI ·
Your next cue,
from NOCUE.
Research
Principled Thoughts for Latent Recursive LLM Systems (opens in a new tab)
arXiv cs.AI ·
Research
An Exact Generate - Transform Decomposition of Small-LLM Team Scaling Across Orchestration Architectures (opens in a new tab)
arXiv cs.AI ·
Research
Mirror-Score: Calibrated, Inference-only Scoring Exposes the Limits of Sequence-compatibility Ranking in D-peptide Design (opens in a new tab)
arXiv cs.AI ·
Research
Right Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT (opens in a new tab)
arXiv cs.AI ·
Research
GeoOutageBench: Benchmarking Ambiguity-aware, Ontology-grounded Geospatiotemporal KGQA for Multimodal Power Outage and Resilience Analysis (opens in a new tab)
arXiv cs.AI ·
Research
Grab a Coffee: Future-Aware Guidance for Discrete Diffusion with Compiled Objectives (opens in a new tab)
arXiv cs.AI ·
Research
Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial Training (opens in a new tab)
arXiv cs.AI ·
Research
Self-discovering RL in the Era of Experience: Is Learning History an Asset or a Burden? (opens in a new tab)
arXiv cs.AI ·